Merge branch 'main' into feat/shortcut-protocol

fix(shortcuts): typed []string named-element binding + nil-Execute mount parity
Two parity edge cases from a follow-up audit: - []<named string> element types (type ID string; field []ID) passed parse validation but panicked at bind via reflect.Convert([]string -> []ID). Build the slice element-by-element with SetString (stringsToSlice) so plain []string, named slice types and named element types all bind correctly. - nil Execute: mountTyped now skips mounting (matching legacy Shortcut.MountWithContext) instead of mounting a command that errors only at invocation, keeping the command tree identical after migration. Adds binder_namedslice_nilexec_test.go (4 cases).
2026-07-06 00:06:28 +08:00 · 2026-06-03 13:50:45 +08:00 · 2026-05-29 15:30:20 +08:00 · 2026-05-29 15:15:09 +08:00 · 2026-05-29 14:55:40 +08:00 · 2026-05-29 11:51:16 +08:00
1454 changed files with 25243 additions and 229206 deletions
--- a/.github/CODEOWNERS
+++ b/.github/CODEOWNERS
@@ -1,30 +0,0 @@
-/internal/ @liangshuo-1
-
-# Last match wins: existing domains below are exempt, only new skills/ entries need review.
-/skills/ @liangshuo-1
-/skills/lark-approval/
-/skills/lark-apps/
-/skills/lark-attendance/
-/skills/lark-base/
-/skills/lark-calendar/
-/skills/lark-contact/
-/skills/lark-doc/
-/skills/lark-drive/
-/skills/lark-event/
-/skills/lark-im/
-/skills/lark-mail/
-/skills/lark-markdown/
-/skills/lark-minutes/
-/skills/lark-okr/
-/skills/lark-openapi-explorer/
-/skills/lark-shared/
-/skills/lark-sheets/
-/skills/lark-skill-maker/
-/skills/lark-slides/
-/skills/lark-task/
-/skills/lark-vc/
-/skills/lark-vc-agent/
-/skills/lark-whiteboard/
-/skills/lark-wiki/
-/skills/lark-workflow-meeting-summary/
-/skills/lark-workflow-standup-report/
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -5,12 +5,13 @@ on:
    branches: [main]
  pull_request:
    branches: [main]
-    types: [opened, synchronize, reopened, edited]
  workflow_dispatch:

 permissions:
  contents: read
  actions: read
+  checks: write
+  pull-requests: write

 jobs:
  # ── Layer 1: Fast Gate ─────────────────────────────────────────────
@@ -71,7 +72,6 @@ jobs:
      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
        with:
          fetch-depth: 0
-          persist-credentials: false
      - uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
        with:
          go-version-file: go.mod
@@ -80,84 +80,10 @@ jobs:
          python-version: '3.x'
      - name: Fetch meta data
        run: python3 scripts/fetch_meta.py
-      - name: Resolve changed-from baseline
-        env:
-          QUALITY_GATE_CHANGED_FROM: ${{ github.event.pull_request.base.sha || github.event.before || 'origin/main' }}
-        run: echo "QUALITY_GATE_CHANGED_FROM=$(bash scripts/resolve-changed-from.sh)" >> "$GITHUB_ENV"
      - name: Run golangci-lint
-        run: go run github.com/golangci/golangci-lint/v2/cmd/golangci-lint@v2.1.6 run --new-from-rev="$QUALITY_GATE_CHANGED_FROM"
+        run: go run github.com/golangci/golangci-lint/v2/cmd/golangci-lint@v2.1.6 run --new-from-rev=origin/main
      - name: Run errs/ lint guards (lintcheck)
-        run: go run -C lint . --changed-from "$QUALITY_GATE_CHANGED_FROM" ..
-
-  script-test:
-    needs: fast-gate
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
-        with:
-          fetch-depth: 0
-          persist-credentials: false
-      - uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
-        with:
-          go-version-file: go.mod
-      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4
-        with:
-          node-version: '22'
-      - name: Run script tests
-        run: make script-test
-
-  deterministic-gate:
-    needs: fast-gate
-    runs-on: ubuntu-latest
-    permissions:
-      contents: read
-      actions: read
-    steps:
-      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
-        with:
-          fetch-depth: 0
-      - uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
-        with:
-          go-version-file: go.mod
-      - uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6
-        with:
-          python-version: '3.x'
-      - name: Fetch meta data
-        run: python3 scripts/fetch_meta.py
-      - name: Resolve changed-from baseline
-        env:
-          QUALITY_GATE_CHANGED_FROM: ${{ github.event.pull_request.base.sha || github.event.before || 'origin/main' }}
-        run: echo "QUALITY_GATE_CHANGED_FROM=$(bash scripts/resolve-changed-from.sh)" >> "$GITHUB_ENV"
-      - name: Write public content metadata
-        if: ${{ github.event_name == 'pull_request' }}
-        env:
-          PR_TITLE: ${{ github.event.pull_request.title }}
-          PR_BODY: ${{ github.event.pull_request.body }}
-          PR_BRANCH: ${{ github.head_ref }}
-        run: |
-          mkdir -p .tmp/quality-gate
-          python3 - <<'PY'
-          import json
-          import os
-
-          with open(".tmp/quality-gate/public-content-metadata.json", "w", encoding="utf-8") as f:
-              json.dump({
-                  "title": os.environ.get("PR_TITLE", ""),
-                  "body": os.environ.get("PR_BODY", ""),
-                  "branch": os.environ.get("PR_BRANCH", ""),
-              }, f)
-              f.write("\n")
-          PY
-      - name: Run CLI deterministic gate
-        run: PUBLIC_CONTENT_METADATA=.tmp/quality-gate/public-content-metadata.json make quality-gate
-      - name: Upload quality gate facts
-        if: ${{ always() && github.event_name == 'pull_request' }}
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
-        with:
-          name: quality-gate-facts-${{ github.event.pull_request.base.sha }}-${{ github.event.pull_request.head.sha }}
-          path: .tmp/quality-gate/facts.json
-          if-no-files-found: error
-          retention-days: 7
+        run: go run -C lint . ..

  coverage:
    needs: fast-gate
@@ -177,7 +103,6 @@ jobs:
          packages=$(go list ./... | grep -v '^github.com/larksuite/cli/tests/cli_e2e$' | grep -v '^github.com/larksuite/cli/tests/cli_e2e/')
          go test -race -coverprofile=coverage.txt -covermode=atomic $packages
      - name: Upload coverage to Codecov
-        if: ${{ github.event_name != 'pull_request' || !github.event.pull_request.head.repo.fork }}
        uses: codecov/codecov-action@3f20e214133d0983f9a10f3d63b0faf9241a3daa # v6
        with:
          files: coverage.txt
@@ -259,7 +184,7 @@ jobs:

  # ── Layer 3: E2E Gate ──────────────────────────────────────────────
  e2e-dry-run:
-    needs: [unit-test, lint, script-test, deterministic-gate]
+    needs: [unit-test, lint]
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
@@ -280,12 +205,9 @@ jobs:
        run: go test -v -count=1 -timeout=5m ./tests/cli_e2e/... -run 'DryRun|Regression'

  e2e-live:
-    needs: [unit-test, lint, script-test, deterministic-gate]
+    needs: [unit-test, lint]
    if: ${{ github.event_name != 'pull_request' || !github.event.pull_request.head.repo.fork }}
    runs-on: ubuntu-latest
-    permissions:
-      contents: read
-      checks: write
    env:
      TEST_BOT1_APP_ID: ${{ secrets.TEST_BOT1_APP_ID }}
      TEST_BOT1_APP_SECRET: ${{ secrets.TEST_BOT1_APP_SECRET }}
@@ -332,9 +254,6 @@ jobs:
  # ── Layer 4: Security & Compliance (parallel with L2-L3) ──────────
  security:
    runs-on: ubuntu-latest
-    permissions:
-      contents: read
-      pull-requests: read
    steps:
      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
        with:
@@ -372,7 +291,7 @@ jobs:
  # ── Results Gate (single required check for branch protection) ─────
  results:
    if: ${{ always() }}
-    needs: [fast-gate, unit-test, lint, script-test, deterministic-gate, coverage, deadcode, e2e-dry-run, e2e-live, security, license-header]
+    needs: [fast-gate, unit-test, lint, coverage, deadcode, e2e-dry-run, e2e-live, security, license-header]
    runs-on: ubuntu-latest
    steps:
      - name: Evaluate results
@@ -384,8 +303,6 @@ jobs:
          echo "| L1 | fast-gate | ${{ needs.fast-gate.result }} |" >> $GITHUB_STEP_SUMMARY
          echo "| L2 | unit-test | ${{ needs.unit-test.result }} |" >> $GITHUB_STEP_SUMMARY
          echo "| L2 | lint | ${{ needs.lint.result }} |" >> $GITHUB_STEP_SUMMARY
-          echo "| L2 | script-test | ${{ needs.script-test.result }} |" >> $GITHUB_STEP_SUMMARY
-          echo "| L2 | deterministic-gate | ${{ needs.deterministic-gate.result }} |" >> $GITHUB_STEP_SUMMARY
          echo "| L2 | coverage | ${{ needs.coverage.result }} |" >> $GITHUB_STEP_SUMMARY
          echo "| L2 | deadcode | ${{ needs.deadcode.result }} |" >> $GITHUB_STEP_SUMMARY
          echo "| L3 | e2e-dry-run | ${{ needs.e2e-dry-run.result }} |" >> $GITHUB_STEP_SUMMARY
@@ -401,8 +318,6 @@ jobs:
            "${{ needs.fast-gate.result }}" \
            "${{ needs.unit-test.result }}" \
            "${{ needs.lint.result }}" \
-            "${{ needs.script-test.result }}" \
-            "${{ needs.deterministic-gate.result }}" \
            "${{ needs.coverage.result }}" \
            "${{ needs.deadcode.result }}" \
            "${{ needs.e2e-dry-run.result }}" \
--- a/.github/workflows/comment-audit.yml
+++ b/.github/workflows/comment-audit.yml
@@ -1,28 +0,0 @@
-name: Comment Audit
-
-on:
-  issue_comment:
-    types: [created, edited]
-  pull_request_review:
-    types: [submitted, edited]
-  pull_request_review_comment:
-    types: [created, edited]
-
-permissions:
-  contents: read
-
-jobs:
-  public-content-comment-audit:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
-        with:
-          persist-credentials: false
-      - uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
-        with:
-          go-version-file: go.mod
-      - name: Post-publication comment audit
-        run: |
-          mkdir -p .tmp/comment-audit
-          cp "$GITHUB_EVENT_PATH" .tmp/comment-audit/event.json
-          go run ./internal/qualitygate/cmd/comment-audit --event .tmp/comment-audit/event.json --kind "$GITHUB_EVENT_NAME"
--- a/.github/workflows/semantic-review.yml
+++ b/.github/workflows/semantic-review.yml
@@ -1,611 +0,0 @@
-name: Semantic Review
-
-on:
-  workflow_run:
-    workflows: ["CI"]
-    types: [completed]
-
-permissions:
-  actions: read
-  contents: read
-
-jobs:
-  pr-quality-summary:
-    if: github.event.workflow_run.event == 'pull_request'
-    runs-on: ubuntu-latest
-    permissions:
-      actions: read
-      contents: read
-      issues: write
-      pull-requests: write
-    steps:
-      - name: Verify workflow run and pull request for summary
-        id: pr
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        with:
-          script: |
-            const run = context.payload.workflow_run;
-            if (run.name !== "CI") throw new Error(`unexpected workflow name: ${run.name}`);
-            let workflowPath = run.path || "";
-            if (!workflowPath) {
-              const workflowId = Number(run.workflow_id || 0);
-              if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
-              const { data: workflow } = await github.rest.actions.getWorkflow({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                workflow_id: workflowId,
-              });
-              workflowPath = workflow.path || "";
-            }
-            if (workflowPath !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflowPath}`);
-            if (run.event !== "pull_request") throw new Error(`unexpected event: ${run.event}`);
-            if (run.repository.id !== context.payload.repository.id) throw new Error("repository id mismatch");
-            if (run.repository.full_name !== context.payload.repository.full_name) throw new Error("repository name mismatch");
-            if (typeof run.head_sha !== "string" || run.head_sha.length !== 40) throw new Error("invalid head sha");
-            const runPRs = Array.isArray(run.pull_requests) ? run.pull_requests : [];
-            if (runPRs.length > 1) {
-              throw new Error(`ambiguous workflow_run pull request bindings: ${runPRs.length}`);
-            }
-            let prNumber = Number(runPRs[0]?.number || 0);
-            const eventBaseSha = runPRs[0]?.base?.sha || "";
-            const eventHeadSha = runPRs[0]?.head?.sha || "";
-            const targetHeadSha = run.head_sha;
-            if (!/^[a-f0-9]{40}$/i.test(targetHeadSha)) throw new Error("invalid PR head sha");
-            if (eventHeadSha && eventHeadSha.toLowerCase() !== targetHeadSha.toLowerCase()) {
-              core.notice("PR quality summary using workflow_run head_sha because workflow_run pull request head differs from the CI run head");
-            }
-
-            const factsArtifactPattern = /^quality-gate-facts-([a-f0-9]{40})-([a-f0-9]{40})$/i;
-            const { data: artifactData } = await github.rest.actions.listWorkflowRunArtifacts({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              run_id: run.id,
-              per_page: 100,
-            });
-            const factsArtifacts = artifactData.artifacts.filter((artifact) => factsArtifactPattern.test(artifact.name));
-            let factsArtifactName = "";
-            let artifactBaseSha = "";
-            let artifactError = "";
-            if (factsArtifacts.length !== 1) {
-              artifactError = `expected exactly one base-bound quality gate facts artifact, got ${factsArtifacts.length}`;
-            } else {
-              factsArtifactName = factsArtifacts[0].name;
-              const [, parsedBaseSha, artifactHeadSha] = factsArtifactName.match(factsArtifactPattern);
-              if (artifactHeadSha.toLowerCase() !== targetHeadSha.toLowerCase()) {
-                artifactError = "facts artifact head sha does not match verified PR head sha";
-                factsArtifactName = "";
-              } else {
-                artifactBaseSha = parsedBaseSha;
-                if (eventBaseSha && parsedBaseSha.toLowerCase() !== eventBaseSha.toLowerCase()) {
-                  core.notice("PR quality summary using facts artifact base because workflow_run pull request base differs from the CI facts artifact base");
-                }
-              }
-            }
-            if (!prNumber) {
-              const { data: associatedPRs } = await github.rest.repos.listPullRequestsAssociatedWithCommit({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                commit_sha: targetHeadSha,
-              });
-              const candidatePRs = associatedPRs.filter((candidate) =>
-                candidate.base?.repo?.id === context.payload.repository.id &&
-                candidate.head?.sha === targetHeadSha
-              );
-              const openCandidatePRs = candidatePRs.filter((candidate) => candidate.state === "open");
-              if (openCandidatePRs.length > 1) {
-                throw new Error(`ambiguous open PRs for workflow_run head ${targetHeadSha}: ${openCandidatePRs.length}`);
-              }
-              if (openCandidatePRs.length === 1) {
-                prNumber = openCandidatePRs[0].number;
-              } else if (candidatePRs.length > 0) {
-                core.notice("PR quality summary skipped: workflow_run target PR is no longer open");
-                core.setOutput("stale", "true");
-                return;
-              }
-            }
-            if (!prNumber) {
-              const candidatePRs = await github.paginate(github.rest.pulls.list, {
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                state: "all",
-                per_page: 100,
-              }).then((prs) => prs.filter((candidate) =>
-                candidate.base?.repo?.id === context.payload.repository.id &&
-                candidate.head?.sha === targetHeadSha
-              ));
-              const openCandidatePRs = candidatePRs.filter((candidate) => candidate.state === "open");
-              if (openCandidatePRs.length > 1) {
-                throw new Error(`ambiguous open PRs from pull list fallback for workflow_run head ${targetHeadSha}: ${openCandidatePRs.length}`);
-              }
-              if (openCandidatePRs.length === 1) {
-                prNumber = openCandidatePRs[0].number;
-              } else if (candidatePRs.length > 0) {
-                core.notice("PR quality summary skipped: workflow_run target PR is no longer open");
-                core.setOutput("stale", "true");
-                return;
-              } else {
-                throw new Error(`expected one open PR from pull list fallback for workflow_run head ${targetHeadSha}, got ${candidatePRs.length}`);
-              }
-            }
-            if (!Number.isInteger(prNumber) || prNumber <= 0) throw new Error("missing pull request binding");
-            const { data: pr } = await github.rest.pulls.get({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              pull_number: prNumber,
-            });
-            if (pr.base.repo.id !== context.payload.repository.id) throw new Error("PR base repo mismatch");
-            if (pr.state !== "open") {
-              core.notice("PR quality summary skipped: workflow_run target PR is no longer open");
-              core.setOutput("stale", "true");
-              return;
-            }
-            if (pr.head.sha !== targetHeadSha) {
-              core.notice("PR quality summary skipped: workflow_run is stale for this PR head");
-              core.setOutput("stale", "true");
-              return;
-            }
-            const baseSha = artifactBaseSha || eventBaseSha || pr.base.sha;
-            if (!/^[a-f0-9]{40}$/i.test(baseSha)) throw new Error("invalid PR base sha");
-            if ((eventBaseSha || artifactBaseSha) && pr.base.sha !== baseSha) {
-              core.notice("PR quality summary skipped: workflow_run is stale for this PR base");
-              core.setOutput("stale", "true");
-              return;
-            }
-            if (artifactError) {
-              core.warning(`quality gate facts artifact binding is unavailable: ${artifactError}`);
-            }
-            core.setOutput("pr_number", String(prNumber));
-            core.setOutput("head_sha", targetHeadSha);
-            core.setOutput("base_sha", baseSha);
-            core.setOutput("run_id", String(run.id));
-            core.setOutput("facts_artifact_name", factsArtifactName);
-            core.setOutput("artifact_error", artifactError);
-            core.setOutput("stale", "false");
-
-      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
-        id: checkout
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        with:
-          ref: ${{ steps.pr.outputs.base_sha }}
-          persist-credentials: false
-
-      - name: Verify summary facts artifact metadata
-        id: artifact
-        if: ${{ steps.pr.outputs.stale != 'true' && steps.pr.outputs.facts_artifact_name != '' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        with:
-          script: |
-            const run = context.payload.workflow_run;
-            const factsArtifactName = "${{ steps.pr.outputs.facts_artifact_name }}";
-            const { data } = await github.rest.actions.listWorkflowRunArtifacts({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              run_id: run.id,
-              per_page: 100,
-            });
-            const artifacts = data.artifacts.filter(a => a.name === factsArtifactName);
-            if (artifacts.length !== 1) throw new Error(`expected exactly one quality-gate-facts artifact, got ${artifacts.length}`);
-            const artifact = artifacts[0];
-            if (artifact.expired) throw new Error("quality-gate-facts artifact expired");
-            if (artifact.size_in_bytes <= 0 || artifact.size_in_bytes > 5 * 1024 * 1024) {
-              throw new Error(`invalid artifact size: ${artifact.size_in_bytes}`);
-            }
-            if (!artifact.digest) throw new Error("facts artifact digest is missing from GitHub API response");
-            core.setOutput("artifact_id", String(artifact.id));
-            core.setOutput("artifact_digest", artifact.digest);
-
-      - name: Download facts artifact zip
-        if: ${{ steps.pr.outputs.stale != 'true' && steps.artifact.outputs.artifact_id != '' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        id: download
-        with:
-          script: |
-            const fs = require("fs");
-            const path = require("path");
-            const artifactId = Number("${{ steps.artifact.outputs.artifact_id }}");
-            if (!Number.isInteger(artifactId) || artifactId <= 0) throw new Error("invalid artifact id");
-            const { data } = await github.rest.actions.downloadArtifact({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              artifact_id: artifactId,
-              archive_format: "zip",
-            });
-            const zipPath = path.join(process.env.RUNNER_TEMP, "quality-gate-facts.zip");
-            fs.writeFileSync(zipPath, Buffer.from(data));
-            core.setOutput("zip_path", zipPath);
-
-      - name: Verify and extract summary facts artifact
-        if: ${{ steps.pr.outputs.stale != 'true' && steps.download.outputs.zip_path != '' }}
-        env:
-          SEMANTIC_REVIEW_BLOCK: ${{ vars.SEMANTIC_REVIEW_BLOCK }}
-          SEMANTIC_REVIEW_DECISION_OUT: decision.json
-          SEMANTIC_REVIEW_MARKDOWN_OUT: semantic-review.md
-        run: node scripts/semantic-review-verify-artifact.js '${{ steps.download.outputs.zip_path }}' facts.json '${{ steps.artifact.outputs.artifact_digest }}'
-
-      - name: Publish PR quality summary
-        if: ${{ always() && steps.pr.outputs.stale != 'true' && steps.checkout.outcome == 'success' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        env:
-          CI_QUALITY_SUMMARY_HEAD_SHA: ${{ steps.pr.outputs.head_sha }}
-          CI_QUALITY_SUMMARY_BASE_SHA: ${{ steps.pr.outputs.base_sha }}
-          CI_QUALITY_SUMMARY_PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
-          CI_QUALITY_SUMMARY_RUN_ID: ${{ steps.pr.outputs.run_id }}
-          CI_QUALITY_SUMMARY_ARTIFACT_ERROR: ${{ steps.pr.outputs.artifact_error }}
-        with:
-          script: |
-            const { publish } = require("./scripts/ci-quality-summary-publish.js");
-            await publish({ github, context, core });
-
-  semantic-review:
-    needs: pr-quality-summary
-    if: always() && github.event.workflow_run.conclusion == 'success' && github.event.workflow_run.event == 'pull_request'
-    runs-on: ubuntu-latest
-    permissions:
-      actions: read
-      checks: write
-      contents: read
-      issues: write
-      pull-requests: write
-    steps:
-      - name: Verify workflow run and pull request
-        id: pr
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        with:
-          script: |
-            const run = context.payload.workflow_run;
-            if (run.name !== "CI") throw new Error(`unexpected workflow name: ${run.name}`);
-            let workflowPath = run.path || "";
-            if (!workflowPath) {
-              const workflowId = Number(run.workflow_id || 0);
-              if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
-              const { data: workflow } = await github.rest.actions.getWorkflow({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                workflow_id: workflowId,
-              });
-              workflowPath = workflow.path || "";
-            }
-            if (workflowPath !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflowPath}`);
-            if (run.event !== "pull_request") throw new Error(`unexpected event: ${run.event}`);
-            if (run.conclusion !== "success") throw new Error(`unexpected conclusion: ${run.conclusion}`);
-            if (run.repository.id !== context.payload.repository.id) throw new Error("repository id mismatch");
-            if (run.repository.full_name !== context.payload.repository.full_name) throw new Error("repository name mismatch");
-            if (typeof run.head_sha !== "string" || run.head_sha.length !== 40) throw new Error("invalid head sha");
-            const runPRs = Array.isArray(run.pull_requests) ? run.pull_requests : [];
-            if (runPRs.length > 1) {
-              throw new Error(`ambiguous workflow_run pull request bindings: ${runPRs.length}`);
-            }
-            let prNumber = Number(runPRs[0]?.number || 0);
-            const eventBaseSha = runPRs[0]?.base?.sha || "";
-            const eventHeadSha = runPRs[0]?.head?.sha || "";
-            const targetHeadSha = run.head_sha;
-            if (!/^[a-f0-9]{40}$/i.test(targetHeadSha)) throw new Error("invalid PR head sha");
-            if (eventHeadSha && eventHeadSha.toLowerCase() !== targetHeadSha.toLowerCase()) {
-              core.notice("semantic review using workflow_run head_sha because workflow_run pull request head differs from the CI run head");
-            }
-
-            const factsArtifactPattern = /^quality-gate-facts-([a-f0-9]{40})-([a-f0-9]{40})$/i;
-            const { data: artifactData } = await github.rest.actions.listWorkflowRunArtifacts({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              run_id: run.id,
-              per_page: 100,
-            });
-            const factsArtifacts = artifactData.artifacts.filter((artifact) => factsArtifactPattern.test(artifact.name));
-            let factsArtifactName = "";
-            let artifactBaseSha = "";
-            let artifactError = "";
-            if (factsArtifacts.length !== 1) {
-              artifactError = `expected exactly one base-bound quality gate facts artifact, got ${factsArtifacts.length}`;
-            } else {
-              factsArtifactName = factsArtifacts[0].name;
-              const [, parsedBaseSha, artifactHeadSha] = factsArtifactName.match(factsArtifactPattern);
-              if (artifactHeadSha.toLowerCase() !== targetHeadSha.toLowerCase()) {
-                artifactError = "facts artifact head sha does not match verified PR head sha";
-                factsArtifactName = "";
-              } else {
-                artifactBaseSha = parsedBaseSha;
-                if (eventBaseSha && parsedBaseSha.toLowerCase() !== eventBaseSha.toLowerCase()) {
-                  core.notice("semantic review using facts artifact base because workflow_run pull request base differs from the CI facts artifact base");
-                }
-              }
-            }
-            if (!prNumber) {
-              const { data: associatedPRs } = await github.rest.repos.listPullRequestsAssociatedWithCommit({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                commit_sha: targetHeadSha,
-              });
-              const candidatePRs = associatedPRs.filter((candidate) =>
-                candidate.base?.repo?.id === context.payload.repository.id &&
-                candidate.head?.sha === targetHeadSha
-              );
-              const openCandidatePRs = candidatePRs.filter((candidate) => candidate.state === "open");
-              if (openCandidatePRs.length > 1) {
-                throw new Error(`ambiguous open PRs for workflow_run head ${targetHeadSha}: ${openCandidatePRs.length}`);
-              }
-              if (openCandidatePRs.length === 1) {
-                prNumber = openCandidatePRs[0].number;
-              } else if (candidatePRs.length > 0) {
-                core.notice("semantic review skipped: workflow_run target PR is no longer open");
-                core.setOutput("stale", "true");
-                return;
-              }
-            }
-            if (!prNumber) {
-              const candidatePRs = await github.paginate(github.rest.pulls.list, {
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                state: "all",
-                per_page: 100,
-              }).then((prs) => prs.filter((candidate) =>
-                candidate.base?.repo?.id === context.payload.repository.id &&
-                candidate.head?.sha === targetHeadSha
-              ));
-              const openCandidatePRs = candidatePRs.filter((candidate) => candidate.state === "open");
-              if (openCandidatePRs.length > 1) {
-                throw new Error(`ambiguous open PRs from pull list fallback for workflow_run head ${targetHeadSha}: ${openCandidatePRs.length}`);
-              }
-              if (openCandidatePRs.length === 1) {
-                prNumber = openCandidatePRs[0].number;
-              } else if (candidatePRs.length > 0) {
-                core.notice("semantic review skipped: workflow_run target PR is no longer open");
-                core.setOutput("stale", "true");
-                return;
-              } else {
-                throw new Error(`expected one open PR from pull list fallback for workflow_run head ${targetHeadSha}, got ${candidatePRs.length}`);
-              }
-            }
-            if (!Number.isInteger(prNumber) || prNumber <= 0) throw new Error("missing pull request binding");
-            const { data: pr } = await github.rest.pulls.get({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              pull_number: prNumber,
-            });
-            if (pr.base.repo.id !== context.payload.repository.id) throw new Error("PR base repo mismatch");
-            if (pr.state !== "open") {
-              core.notice("semantic review skipped: workflow_run target PR is no longer open");
-              core.setOutput("stale", "true");
-              return;
-            }
-            if (!pr.head.repo) {
-              core.notice("semantic review skipped: workflow_run target PR head repository is unavailable");
-              core.setOutput("stale", "true");
-              return;
-            }
-            if (pr.head.sha !== targetHeadSha) {
-              core.notice("semantic review skipped: workflow_run is stale for this PR head");
-              core.setOutput("stale", "true");
-              return;
-            }
-            const baseSha = artifactBaseSha || eventBaseSha || pr.base.sha;
-            if (!/^[a-f0-9]{40}$/i.test(baseSha)) throw new Error("invalid PR base sha");
-            if ((eventBaseSha || artifactBaseSha) && pr.base.sha !== baseSha) {
-              core.notice("semantic review skipped: workflow_run is stale for this PR base");
-              core.setOutput("stale", "true");
-              return;
-            }
-            if (artifactError) {
-              core.warning(`semantic review facts artifact binding is unavailable: ${artifactError}`);
-            }
-            core.setOutput("pr_number", String(prNumber));
-            core.setOutput("head_sha", targetHeadSha);
-            core.setOutput("base_sha", baseSha);
-            core.setOutput("head_owner", pr.head.repo.owner.login);
-            core.setOutput("head_repo", pr.head.repo.name);
-            core.setOutput("head_repo_id", String(pr.head.repo.id));
-            core.setOutput("head_is_base_repo", pr.head.repo.id === context.payload.repository.id ? "true" : "false");
-            core.setOutput("run_id", String(run.id));
-            core.setOutput("facts_artifact_name", factsArtifactName);
-            core.setOutput("artifact_error", artifactError);
-            core.setOutput("stale", "false");
-
-      - uses: actions/checkout@93cb6efe18208431cddfb8368fd83d5badbf9bfd # v5
-        id: checkout
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        with:
-          ref: ${{ steps.pr.outputs.base_sha }}
-          persist-credentials: false
-
-      - name: Publish pre-checkout semantic review failure
-        if: ${{ failure() && steps.pr.outputs.stale != 'true' && steps.checkout.outcome != 'success' && steps.pr.outputs.head_sha != '' && steps.pr.outputs.pr_number != '' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        env:
-          SEMANTIC_REVIEW_BLOCK: ${{ vars.SEMANTIC_REVIEW_BLOCK }}
-          SEMANTIC_REVIEW_HEAD_SHA: ${{ steps.pr.outputs.head_sha }}
-          SEMANTIC_REVIEW_BASE_SHA: ${{ steps.pr.outputs.base_sha }}
-          SEMANTIC_REVIEW_PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
-          SEMANTIC_REVIEW_RUN_ID: ${{ steps.pr.outputs.run_id }}
-        with:
-          script: |
-            const runtimeBlockMode = process.env.SEMANTIC_REVIEW_BLOCK === "true";
-            const pr = Number(process.env.SEMANTIC_REVIEW_PR_NUMBER || 0);
-            const headSha = process.env.SEMANTIC_REVIEW_HEAD_SHA || "";
-            const baseSha = process.env.SEMANTIC_REVIEW_BASE_SHA || "";
-            if (!Number.isInteger(pr) || pr <= 0 || !/^[a-f0-9]{40}$/i.test(headSha) || !/^[a-f0-9]{40}$/i.test(baseSha)) {
-              throw new Error("missing verified semantic review target");
-            }
-            const { data: pull } = await github.rest.pulls.get({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              pull_number: pr,
-            });
-            if (pull.state !== "open") {
-              core.notice("semantic review skipped infrastructure failure check: PR is no longer open");
-              return;
-            }
-            if (pull.head.sha !== headSha) {
-              core.notice("semantic review skipped infrastructure failure check: PR head changed");
-              return;
-            }
-            if (pull.base.sha !== baseSha) {
-              core.notice("semantic review skipped infrastructure failure check: PR base changed");
-              return;
-            }
-            if (pull.base.repo.id !== context.payload.repository.id) {
-              throw new Error("PR base repo mismatch before infrastructure failure check");
-            }
-            await github.rest.checks.create({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              name: runtimeBlockMode ? "semantic-review/result" : "semantic-review/observe",
-              head_sha: headSha,
-              status: "completed",
-              conclusion: runtimeBlockMode ? "failure" : "neutral",
-              output: {
-                title: "Semantic review infrastructure failure",
-                summary: "Semantic review could not checkout the verified base commit. Inspect the workflow logs before relying on semantic review output.",
-              },
-            });
-
-      - uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        with:
-          go-version-file: go.mod
-
-      - name: Verify semantic facts artifact metadata
-        id: artifact
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        with:
-          script: |
-            const run = context.payload.workflow_run;
-            const factsArtifactName = "${{ steps.pr.outputs.facts_artifact_name }}";
-            if (!/^quality-gate-facts-[a-f0-9]{40}-[a-f0-9]{40}$/i.test(factsArtifactName)) {
-              throw new Error("missing verified facts artifact binding");
-            }
-            const { data } = await github.rest.actions.listWorkflowRunArtifacts({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              run_id: run.id,
-              per_page: 100,
-            });
-            const artifacts = data.artifacts.filter(a => a.name === factsArtifactName);
-            if (artifacts.length !== 1) throw new Error(`expected exactly one quality-gate-facts artifact, got ${artifacts.length}`);
-            const artifact = artifacts[0];
-            if (artifact.expired) throw new Error("quality-gate-facts artifact expired");
-            if (artifact.size_in_bytes <= 0 || artifact.size_in_bytes > 5 * 1024 * 1024) {
-              throw new Error(`invalid artifact size: ${artifact.size_in_bytes}`);
-            }
-            if (!artifact.digest) throw new Error("facts artifact digest is missing from GitHub API response");
-            core.setOutput("artifact_id", String(artifact.id));
-            core.setOutput("artifact_digest", artifact.digest);
-
-      - name: Download facts artifact zip
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        id: download
-        with:
-          script: |
-            const fs = require("fs");
-            const path = require("path");
-            const artifactId = Number("${{ steps.artifact.outputs.artifact_id }}");
-            if (!Number.isInteger(artifactId) || artifactId <= 0) throw new Error("invalid artifact id");
-            const { data } = await github.rest.actions.downloadArtifact({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              artifact_id: artifactId,
-              archive_format: "zip",
-            });
-            const zipPath = path.join(process.env.RUNNER_TEMP, "quality-gate-facts.zip");
-            fs.writeFileSync(zipPath, Buffer.from(data));
-            core.setOutput("zip_path", zipPath);
-
-      - name: Verify and extract semantic facts artifact
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        env:
-          SEMANTIC_REVIEW_BLOCK: ${{ vars.SEMANTIC_REVIEW_BLOCK }}
-          SEMANTIC_REVIEW_DECISION_OUT: decision.json
-          SEMANTIC_REVIEW_MARKDOWN_OUT: semantic-review.md
-        run: node scripts/semantic-review-verify-artifact.js '${{ steps.download.outputs.zip_path }}' facts.json '${{ steps.artifact.outputs.artifact_digest }}'
-
-      - name: Download PR semantic waiver config
-        id: waiver_config
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        env:
-          SEMANTIC_REVIEW_HEAD_SHA: ${{ steps.pr.outputs.head_sha }}
-          SEMANTIC_REVIEW_HEAD_OWNER: ${{ steps.pr.outputs.head_owner }}
-          SEMANTIC_REVIEW_HEAD_REPO: ${{ steps.pr.outputs.head_repo }}
-          SEMANTIC_REVIEW_HEAD_IS_BASE_REPO: ${{ steps.pr.outputs.head_is_base_repo }}
-        with:
-          script: |
-            const fs = require("fs");
-            const path = require("path");
-            const headSha = process.env.SEMANTIC_REVIEW_HEAD_SHA || "";
-            if (!/^[a-f0-9]{40}$/i.test(headSha)) {
-              throw new Error("missing verified semantic review target");
-            }
-            const headOwner = process.env.SEMANTIC_REVIEW_HEAD_OWNER || "";
-            const headRepo = process.env.SEMANTIC_REVIEW_HEAD_REPO || "";
-            if (!headOwner || !headRepo) {
-              throw new Error("missing verified semantic review head repository");
-            }
-            const waiverPath = "internal/qualitygate/config/semantic/waivers.txt";
-            const outPath = path.join(process.env.RUNNER_TEMP, "semantic-review-waivers.txt");
-            const headIsBaseRepo = process.env.SEMANTIC_REVIEW_HEAD_IS_BASE_REPO === "true";
-            if (!headIsBaseRepo) {
-              core.notice("fork PR semantic waiver config is ignored");
-              core.setOutput("path", "");
-              return;
-            }
-            let content = "";
-            try {
-              const { data } = await github.rest.repos.getContent({
-                owner: headOwner,
-                repo: headRepo,
-                path: waiverPath,
-                ref: headSha,
-              });
-              if (Array.isArray(data) || data.type !== "file" || data.encoding !== "base64") {
-                throw new Error(`${waiverPath} is not a base64 file at PR head`);
-              }
-              if (data.size > 256 * 1024) {
-                throw new Error(`${waiverPath} is too large: ${data.size} bytes`);
-              }
-              content = Buffer.from(data.content, "base64").toString("utf8");
-            } catch (err) {
-              if (err.status !== 404) {
-                throw err;
-              }
-            }
-            fs.writeFileSync(outPath, content);
-            core.setOutput("path", outPath);
-
-      - name: Run semantic review
-        id: semantic
-        if: ${{ steps.pr.outputs.stale != 'true' }}
-        env:
-          ARK_API_KEY: ${{ secrets.ARK_API_KEY }}
-          ARK_BASE_URL: ${{ vars.ARK_BASE_URL }}
-          ARK_MODEL: ${{ vars.ARK_MODEL }}
-          ARK_TIMEOUT_SECONDS: ${{ vars.ARK_TIMEOUT_SECONDS }}
-          SEMANTIC_REVIEW_BLOCK: ${{ vars.SEMANTIC_REVIEW_BLOCK }}
-        run: |
-          args=(
-            --repo .
-            --facts facts.json
-            --decision-out decision.json
-            --markdown-out semantic-review.md
-          )
-          if [ -n "${{ steps.waiver_config.outputs.path }}" ]; then
-            args+=(--waivers-file '${{ steps.waiver_config.outputs.path }}')
-          fi
-          if [ "$SEMANTIC_REVIEW_BLOCK" = "true" ]; then
-            args+=(--block)
-          fi
-          go run ./internal/qualitygate/cmd/semantic-review "${args[@]}"
-
-      - name: Publish semantic review
-        if: ${{ always() && steps.pr.outputs.stale != 'true' && steps.checkout.outcome == 'success' }}
-        uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
-        env:
-          SEMANTIC_REVIEW_BLOCK: ${{ vars.SEMANTIC_REVIEW_BLOCK }}
-          SEMANTIC_REVIEW_HEAD_SHA: ${{ steps.pr.outputs.head_sha }}
-          SEMANTIC_REVIEW_BASE_SHA: ${{ steps.pr.outputs.base_sha }}
-          SEMANTIC_REVIEW_PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
-          SEMANTIC_REVIEW_RUN_ID: ${{ steps.pr.outputs.run_id }}
-        with:
-          script: |
-            const { publish } = require("./scripts/semantic-review-publish.js");
-            await publish({ github, context, core });
--- a/.gitignore
+++ b/.gitignore
@@ -7,11 +7,6 @@ bin/
 # Node
 node_modules/

-# Python (skill-bundled helper scripts)
-__pycache__/
-*.py[cod]
-*$py.class
-

 # OS
 .DS_Store
@@ -40,8 +35,6 @@ tests/mail/reports/
 # Generated / test artifacts
 .hammer/
 .lark-slides/
-/notes/
-/minutes/
 internal/registry/meta_data.json
 cmd/api/download.bin
 app.log
@@ -51,4 +44,3 @@ app.log
 cover*.out

 lark-env.sh
-/automations/
--- a/.golangci.yml
+++ b/.golangci.yml
@@ -29,11 +29,11 @@ linters:
    - unused          # checks for unused constants, variables, functions and types
    - depguard        # blocks forbidden package imports
    - forbidigo       # forbids specific function calls
-    - errorlint       # enforces error wrapping (%w) and errors.Is/As over == and type asserts

    # To enable later after fixing existing issues:
    # - errcheck      # checks for unchecked errors
    # - errname       # checks that error types are named XxxError
+    # - errorlint     # checks error wrapping best practices
    # - gosec         # security-oriented linter
    # - misspell      # finds commonly misspelled English words
    # - staticcheck   # comprehensive static analysis
@@ -49,49 +49,39 @@ linters:
          - gocritic
          - depguard
          - forbidigo
-          - errorlint # tests legitimately do identity (==) and concrete type-assert checks
-      # forbidigo runs repo-wide (minus the boundaries below) so errs-no-bare-wrap
-      # has no gap. The framework bans (os/vfs, raw HTTP, fmt.Print, filepath,
-      # log) stay scoped to shortcuts/ + internal/ + config/auth/service via the
-      # next rule; elsewhere only errs-no-bare-wrap fires.
-      - path-except: (shortcuts/|internal/|cmd/|events/)
-        linters:
-          - forbidigo
+      # Paths that run forbidigo. Add an entry when a path joins one of
+      # the rules below.
      - path-except: (shortcuts/|internal/|cmd/auth/|cmd/config/|cmd/service/)
-        text: (vfs|IOStreams|ctx\.Out|shortcuts-no-raw-http|filepath functions|os\.Exit|structured error return)
        linters:
          - forbidigo
      - path: internal/vfs/
        linters:
          - forbidigo
-      # internal/gen build-time generators (standalone `package main` run via
-      # go:generate) are not shortcut runtime code — no ctx/runtime/framework —
-      # so the shortcut forbidigo bans don't apply. Going "compliant" is also
-      # impossible here: a structured error return needs os.Exit (also banned),
-      # and the vfs.Xxx() alternative is blocked by depguard shortcuts-no-vfs.
-      - path: shortcuts/.*/internal/gen/
-        linters:
-          - forbidigo
-      # internal/qualitygate/cmd contains standalone CI tools. Their main
-      # entrypoints legitimately own process exit codes and stdio, matching the
-      # old tools/ layout before these packages moved under internal/.
-      - path: internal/qualitygate/cmd/[^/]+/main\.go$
-        linters:
-          - forbidigo
      # shortcuts-no-raw-http is shortcuts-only; internal/ wraps raw HTTP
      # for the client / credential layer.
      - path-except: shortcuts/
        text: shortcuts-no-raw-http
        linters:
          - forbidigo
-      # errs-no-bare-wrap enforced across every command/wire boundary by
-      # structural prefix, so any future business domain or command is covered
-      # without editing an allowlist. Genuine intermediate wraps inside these
-      # paths use //nolint:forbidigo with a reason.
-      - path-except: (cmd/|shortcuts/|events/)
+      # errs-typed-only enforced on paths already migrated to errs.NewXxxError.
+      # Add a path when its migration is complete.
+      - path-except: (internal/auth/|internal/errcompat/|internal/errclass/|internal/client/|internal/cmdutil/factory\.go|cmd/auth/|cmd/config/|cmd/service/|shortcuts/common/mcp_client\.go|shortcuts/calendar/helpers\.go|shortcuts/drive/)
+        text: errs-typed-only
+        linters:
+          - forbidigo
+      # errs-no-bare-wrap enforced on paths fully migrated to typed final
+      # errors. Scoped separately from errs-typed-only because cmd/auth/,
+      # cmd/config/ still have residual fmt.Errorf and must not be caught.
+      - path-except: (shortcuts/drive/|shortcuts/calendar/helpers\.go|shortcuts/common/mcp_client\.go)
        text: errs-no-bare-wrap
        linters:
          - forbidigo
+      # errs-no-legacy-helper is drive-only: the shared helpers it bans are
+      # still used by other domains until their later migration phase.
+      - path-except: (shortcuts/drive/)
+        text: errs-no-legacy-helper
+        linters:
+          - forbidigo

  settings:
    depguard:
@@ -110,6 +100,24 @@ linters:
                Use runtime.FileIO() for file operations or runtime.ValidatePath() for path validation.
    forbidigo:
      forbid:
+        # ── legacy output.Err* helpers banned on migrated paths ──
+        # output.ErrBare is intentionally not listed — it is the predicate-
+        # command silent-exit signal, outside the typed envelope contract.
+        - pattern: output\.(ErrValidation|ErrAuth|ErrNetwork|ErrAPI|ErrWithHint|Errorf)\b
+          msg: >-
+            [errs-typed-only] use errs.NewXxxError(...) builder
+            (see errs/types.go).
+        # ── legacy shared error helpers banned on drive ──
+        # These helpers internally produce legacy output.Err* shapes, so they
+        # are invisible to the errs-typed-only ban above. Drive has migrated its
+        # calls to typed errs.* (drive-local driveInputStatError / driveSaveError);
+        # this prevents reintroduction. Other domains still use the shared
+        # helpers (migrated globally in a later phase), so this is drive-scoped.
+        - pattern: (common\.FlagErrorf|common\.WrapInputStatError|common\.WrapSaveErrorByCategory)\b
+          msg: >-
+            [errs-no-legacy-helper] these shared helpers emit legacy output.Err*
+            shapes. Use the typed errs.NewXxxError builders or the drive-local
+            driveInputStatError / driveSaveError helpers (shortcuts/drive/drive_errors.go).
        # ── bare error wraps banned on fully-typed paths ──
        - pattern: (fmt\.Errorf|errors\.New)\b
          msg: >-
--- a/.goreleaser.yml
+++ b/.goreleaser.yml
@@ -17,7 +17,6 @@ builds:
    goarch:
      - amd64
      - arm64
-      - riscv64

 archives:
  - name_template: "lark-cli-{{ .Version }}-{{ .Os }}-{{ .Arch }}"
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -11,7 +11,7 @@

 ```bash
 make build          # Build (runs fetch_meta first)
-make unit-test      # Required before PR (runs with -race where supported, e.g. amd64/arm64)
+make unit-test      # Required before PR (runs with -race)
 make test           # Full: vet + unit + integration
 ```

@@ -75,31 +75,7 @@ The one rule to internalize: **every error message you write will be parsed by a

 ### Structured errors in commands

-Command-facing failures must be typed `errs.*` errors — never the legacy `output.Err*` helpers and never a final bare `fmt.Errorf`. AI agents parse the stderr envelope's `type` / `subtype` / `param` / `hint` fields to decide their next action; the full taxonomy lives in `errs/ERROR_CONTRACT.md`.
-
-Picking a constructor:
-
-| Failure | Constructor |
-|---------|-------------|
-| User flag/arg fails validation | `errs.NewValidationError(errs.SubtypeInvalidArgument, ...).WithParam("--flag")` |
-| Valid request, wrong system state | `errs.NewValidationError(errs.SubtypeFailedPrecondition, ...).WithHint(...)` |
-| Lark API returned `code != 0` | `runtime.CallAPITyped` (shortcuts) / `errclass.BuildAPIError` (raw responses) — never hand-build |
-| Network / transport failure | `errs.NewNetworkError(errs.SubtypeNetworkTransport, ...)` |
-| Local file I/O failure | `errs.NewInternalError(errs.SubtypeFileIO, ...)` — validate the path first (`validate.SafeInputPath` / `SafeOutputPath`) and use `vfs.*` |
-| Unclassified lower-layer error as final | `errs.NewInternalError(errs.SubtypeUnknown, ...).WithCause(err)` |
-| Lower layer already returned a typed error | pass it through unchanged — re-wrapping downgrades its classification |
-
-Signatures that are easy to guess wrong:
-
- `runtime.CallAPITyped(method, url string, params map[string]interface{}, data interface{}) (map[string]interface{}, error)` — it performs the HTTP request itself and classifies `code != 0` into a typed error; just return the error it gives you.
- Typed pass-through check: `if _, ok := errs.ProblemOf(err); ok { return err }` — `ProblemOf` returns `(*errs.Problem, bool)`, not a nilable pointer.
- `.WithParam` exists only on `*errs.ValidationError`. `InternalError` / `NetworkError` have no param field — file or endpoint context goes in the message or `.WithHint(...)`.
-
-`forbidigo` + `lint/errscontract` reject the legacy `output.Err*` helpers, bare final `fmt.Errorf` / `errors.New`, and legacy envelope literals on migrated paths. Beyond what lint catches, three authoring conventions apply:
-
- Preserve the underlying error with `.WithCause(err)` so `errors.Is` / `errors.Unwrap` keep working.
- `param` names only the user input that actually failed. Recovery guidance goes in `.WithHint(...)`; machine-readable recovery fields (`missing_scopes`, `log_id`) carry server/system ground truth only — never caller-side guesses.
- Error-path tests assert typed metadata via `errs.ProblemOf` (`category` / `subtype` / `param`) and cause preservation, not message substrings alone.
+`RunE` functions must return `output.Errorf` / `output.ErrWithHint` — never bare `fmt.Errorf`. AI agents parse stderr as JSON; bare errors break this contract.

 ### stdout is data, stderr is everything else

--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,350 +2,6 @@

 All notable changes to this project will be documented in this file.

-## [v1.0.62] - 2026-07-01
-
-### Features
-
- **vc**: Add meeting message send shortcut (#1643)
- **doc**: Add document word statistics helper (#1697)
- **cli**: Interactive upgrade prompt for bare `lark-cli` invocation (#1498)
- **install**: Fail closed when `checksums.txt` is missing during install (#1503)
-
-### Bug Fixes
-
- **drive**: Improve batch failure handling for push/pull/sync (#1703)
- **base**: Support JSON array input for field create (#1661)
- **task**: Expose completion state in `my tasks` output (#1641)
- **cli**: Reduce public content credential false positives (#1700)
-
-## [v1.0.61] - 2026-06-30
-
-### Features
-
- **apps**: Add `db`, `file`, `openapi-key` and observability shortcuts (#1596)
- **identity**: Add `whoami` command showing effective identity (#1666)
- **docs**: Add reference map flags (#1547)
-
-### Bug Fixes
-
- **identity**: Correct identity diagnosis under external credential providers (#1693)
- **cli**: Harden git credential error handling (#1676)
-
-### Documentation
-
- **doc**: Guide document copy skill usage (#1673)
- **doc**: Fix lark-doc media token examples (#1662)
-
-## [v1.0.60] - 2026-06-29
-
-### Features
-
- **affordance**: Per-command usage guidance system with markdown source (#1565)
- **event**: Support VC meeting lifecycle events (#1632)
- **sheets**: Use `office_sheet_file` parent_type for imported office spreadsheets (#1606)
- **authorization**: Expand lark-shared auth guidance and assert clean logout JSON (#1598)
- **transport**: Add `LARK_CLI_NO_PROXY_WARN` to silence proxy warning (#1647)
-
-### Bug Fixes
-
- **install**: Load `@clack/prompts` via dynamic import to avoid `ERR_REQUIRE_ESM` (#1652)
-
-### Tests
-
- **doc**: Derive fetch test flag defaults from `v2FetchFlags` (#1428)
-
-### Build
-
- **ci**: Reduce public content false positives
-
-## [v1.0.59] - 2026-06-26
-
-### Features
-
- **slides**: Add `+replace-pages` and `xml get` shortcuts, and expose the presentation URL (#1585)
- **minutes**: Support speaker list and no-Lark speaker replace (#1594)
- **calendar/vc/minutes**: Optimize and extend calendar, vc, minutes, and note shortcuts and skills (#1571)
-
-### Bug Fixes
-
- **docs**: Hide docs `api-version` compat flag (#1580)
-
-## [v1.0.58] - 2026-06-25
-
-### Features
-
- **sheets**: Typed table I/O and error contract, workbook import/export, and skill refresh (#1355)
- **base**: Add Base URL and title resolve shortcuts (#1338)
- **drive**: Add `+member-add` shortcut with wiki space member collection collaborator support (#1204)
- **doc**: Support `create` title option (#1536)
- **doc**: Add `im-markdown` output format for doc fetch (#1550)
- **whiteboard**: Export whiteboard as SVG and update whiteboard via SVG (#1559)
- **card**: Support `card.action.trigger` event with auto-fetched card content (#1528)
- **task**: Add task event consumer (#1510)
-
-### Bug Fixes
-
- **doc**: Prefix docs resource shortcuts (#1564)
- **binding**: Skip unix mode audit on Windows (#1525)
-
-### Documentation
-
- **approval**: Sync approval skill for meta API commands (#1499)
- **doc**: Restore lark-doc style requirements (#1579)
- **im**: Document `chat.nickname` get/update/delete (#1378)
- **im**: Clarify audio message opus requirement (#1271)
-
-### Build
-
- **ci**: Add public content safeguards and reduce false positives
-
-## [v1.0.57] - 2026-06-23
-
-### Features
-
- **slides**: Add `+screenshot` to capture slide page images (or render a single `<slide>` XML snippet), returning the local file path instead of Base64 (#1358)
- **base**: Support record comments (#1043)
- **search**: Surface search API notices (#1413)
-
-### Bug Fixes
-
- **mail**: Resolve folder/label filter once per `+triage list` call (#1512)
- **meta**: Backfill enum value descriptions from options (#1541)
- **cli**: Add missing CLI headers for git credential helper (#1539)
-
-### Documentation
-
- **doc**: Refine rich block, path, and block ID guidance (#1508)
- **mail**: Trim lark-mail skill context (#1527)
- **drive**: Add permission governance workflow guidance (#1292)
-
-### Build
-
- **ci**: Bind semantic review to workflow run head (#1551)
-
-## [v1.0.56] - 2026-06-18
-
-### Features
-
- **apps**: Add `+session-messages-list` for session turn reply messages (#1402)
-
-### Bug Fixes
-
- **api**: Align API success envelopes (#1489)
- **base**: Reject out-of-range pagination flags (#1495)
-
-### Refactor
-
- Retire legacy error envelopes and enforce typed contract (#1449)
-
-### Documentation
-
- **skills**: Soften lark-doc style guidance (#1463)
-
-### Build
-
- Add CI quality gate with semantic review
-
-## [v1.0.55] - 2026-06-16
-
-### Features
-
- **vc**: Support agent meeting event workflows (#1483)
- **drive**: Support exporting Base structure snapshots (#1481)
- **doc**: Add docx cover resource commands (#1468)
- **doc**: Support `lang` for docx fetch v2 (#1459)
- **event**: Optimize subscription precheck, links, and consumer guard (#1447)
-
-### Bug Fixes
-
- **drive**: Validate drive import folder target (#1485)
-
-## [v1.0.54] - 2026-06-15
-
-### Features
-
- **mail**: Auto-attach default signature on send/reply/forward (#1415)
- **drive**: Support `original_creator_ids` filter in search (#1046)
- **cli**: Simplify proxy plugin warning and gate it on TTY (#1448)
-
-### Bug Fixes
-
- **doc**: Fix docs fetch and update ergonomics (#1466)
- **vfs**: Reject blank local paths (#1460)
- **vfs**: Reject Windows absolute paths cross-platform (#1401)
- **event**: Clarify remote bus blocker recovery (#1454)
-
-### Refactor
-
- Converge command pipelines onto a typed metadata model + catalog (#1191)
-
-### Documentation
-
- **im**: Document `@mention` format per message type (text/post/card) (#1419)
- **doc**: Clarify lark-doc create title guidance (#1474)
- **skills**: Add rename prompt for import without `--name` (#1461)
- **apps**: Drop Miaoda brand word from apps command help text (#1399)
-
-## [v1.0.53] - 2026-06-12
-
-### Features
-
- **auth**: Revoke user tokens server-side on `auth logout` (#1434)
- **auth**: Add `--json` flag support to auth subcommands (#1431)
- **token**: Mint TAT via unified OAuth v3 Token Endpoint (#1408)
- **note**: Split note into a dedicated domain with `+detail` and `+transcript` flows (#1345, #1417, #1435)
- **im**: Unify sort flags into `--sort` field and `--order` direction (#1302)
-
-### Bug Fixes
-
- **apps**: Read release error_logs from `data.error_logs` in `+release-get` (#1436)
-
-### Documentation
-
- **skills**: Optimize whiteboard skill (#1371)
- **skills**: Optimize okr skill (#1368)
-
-## [v1.0.52] - 2026-06-11
-
-### Features
-
- **events**: Per-resource subscription identity + Match hook (#1185)
- **apps**: Emit typed error envelopes across the apps domain (#1288)
- **wiki**: Emit typed error envelopes across the wiki domain (#1350)
- **im**: Add `--chat-modes` filter to chat search (#1317)
- **apps**: Exclude `.git` directory from `+html-publish` package (#1396)
- **build**: Support riscv64 prebuilt binaries in release and install pipeline
-
-### Bug Fixes
-
- **apps**: Support git credential dry-run (#1390)
- **whiteboard**: Fix parsing empty whiteboard content (#1391)
- **build**: Make `-race` flag arch-conditional to support riscv64
-
-### Documentation
-
- **im**: Document `chat.user_setting` batch_query/batch_update (#1339)
- **im**: Document `chat.managers` and `chat.moderation` API resources (#1294)
- **skills**: Optimize lark-drive skill routing (#1284)
- **skills**: Expand cite user guidance and fix typos (#1394)
-
-## [v1.0.51] - 2026-06-10
-
-### Features
-
- **apps**: Support multi dev modes (#1175)
- **im**: Complete audio/post rendering and add opt-in `--download-resources` (#1245)
- **base**: Configure initial base table schema (#1377)
- **vc**: Add recording event support (#1369)
- **minutes**: Replace words for transcript (#1372)
- **markdown**: Emit typed error envelopes across the markdown domain (#1347)
- **sheets**: Emit typed error envelopes across the sheets domain (#1348)
- **slides**: Emit typed error envelopes across the slides domain (#1349)
-
-### Documentation
-
- **skills**: Warn about `@file` absolute path restriction in lark-doc skills (#1375)
- **skills**: Remove unsupported ⚠️ from callout emoji list (#1374)
-
-## [v1.0.50] - 2026-06-09
-
-### Features
-
- **doc**: Emit typed error envelopes across the doc domain (#1346)
- **event**: Emit typed error envelopes across the event domain (#1289)
- **contact**: Emit typed error envelopes across the contact domain (#1287)
- **sheets**: Guard `+csv-put --csv` against a path passed without `@` (#1337)
- **cli**: Adjust agent timeout hint output conditions (#1328)
-
-### Bug Fixes
-
- **drive**: Add `@file`/stdin support to `+add-comment --content` (#1343)
- **slides**: Build create URL locally instead of drive metas call (#1329)
- **cli**: Clarify `--block-id` supports comma-separated batch delete in help text (#1336)
-
-### Documentation
-
- **doc**: Replace append with `block_insert_after` in skeleton workflow guidance (#1340)
- **doc**: Document `<folder-manager>` resource block (#1168)
- **drive**: Add drive comment location guidance (#1258)
-
-## [v1.0.49] - 2026-06-08
-
-### Features
-
- **events**: Add whiteboard event domain with per-board subscription (#1265)
- **im**: Support feed group (#1102)
- **im**: Add feed shortcut create, list, and remove shortcuts (#1273)
- **im**: Format feed group error handling (#1308)
- **im**: Return typed error envelopes across the im domain (#1230)
- **base**: Emit typed error envelopes across the base domain (#1248)
- **calendar**: Emit typed error envelopes across the calendar domain (#1232)
- **task**: Emit typed error envelopes across the task domain (#1231)
- **okr,whiteboard**: Emit typed error envelopes across both domains (#1236)
- **minutes,vc**: Emit typed error envelopes across both domains (#1234)
- **markdown**: Harden create upload failures (#1325)
- **drive**: Harden inspect shortcut failures (#1324)
- **slides**: Add IconPark lookup for Lark slides (#1123)
- **doc**: Remove docs v1 API (#1291)
- **cli**: Add `skills` command to read embedded skill content (#1318)
- **cli**: Fetch official skills index (#1301)
- **shared**: Document relative-path-only file arguments (#1319)
- **scopes**: Clear `recommend.allow` scope auto-approve overrides (#1272)
- **shortcuts**: Check shortcut example commands against the live CLI tree (#1244)
-
-### Bug Fixes
-
- **events**: Keep bounded event consume runs alive after stdin EOF (#1285)
- **drive**: Use docs secure label read scope (#1281)
-
-### Documentation
-
- **approval**: Restructure skill with intent table and scope boundaries (#1307)
- **skills**: Tighten drive and markdown guardrails (#1326)
- **skills**: Optimize calendar, vc, and minutes skill guidance (#1269)
- **markdown**: Add markdown domain template (#1293)
- **markdown**: Improve lark-markdown skill guidance (#1279)
- **doc**: Improve lark-doc skill guidance (#1283)
- **wiki**: Optimize skill guidance and routing boundaries (#1275)
- **slides**: Tighten routing/boundary and reconcile in-slide whiteboard (#1169)
-
-## [v1.0.48] - 2026-06-04
-
-### Features
-
- **mail**: Preserve mailbox context in `+triage` output for public mailboxes (#1238)
- **contact**: Add contact skill domain guidance (#1144)
-
-### Bug Fixes
-
- **skills**: Use JSON skills list during update (#1251)
-
-### Documentation
-
- **drive**: Refine lark-drive knowledge organize workflow (#1253)
- **vc-agent**: Require explicit leave request (#1260)
- **slides**: Add whiteboard element documentation and improve slide guidance (#1029)
-
-## [v1.0.47] - 2026-06-03
-
-### Features
-
- **sheets**: Add spec-driven shortcut package with backward-compatible wrapper (#1220)
- **base**: Add base block shortcuts (#1044)
- **im**: Complete card message format (#1198)
- **im**: Improve markdown guidance for messages (#1237)
- **vc**: Forward invite call-id on meeting join (#1243)
- **drive**: Emit typed error envelopes across the drive domain (#1205)
- **common**: Emit typed validation errors from shared shortcut pre-checks (#1242)
- **mail**: Validate `message_ids` in `+messages` before batch get (#1202)
- **wiki**: Support `appid` member type (#1235)
- **cli**: Add `--json` flag as no-op alias for `--format json` (#1104)
- **config**: Validate credentials after `config init` (#1151)
-
-### Bug Fixes
-
- **skills**: Recover empty fallback for skills to update (#1233)
-
 ## [v1.0.46] - 2026-06-02

 ### Features
@@ -1333,22 +989,6 @@ Bundled AI agent skills for intelligent assistance:
 - Bilingual documentation (English & Chinese).
 - CI/CD pipelines: linting, testing, coverage reporting, and automated releases.

-[v1.0.62]: https://github.com/larksuite/cli/releases/tag/v1.0.62
-[v1.0.61]: https://github.com/larksuite/cli/releases/tag/v1.0.61
-[v1.0.60]: https://github.com/larksuite/cli/releases/tag/v1.0.60
-[v1.0.59]: https://github.com/larksuite/cli/releases/tag/v1.0.59
-[v1.0.58]: https://github.com/larksuite/cli/releases/tag/v1.0.58
-[v1.0.57]: https://github.com/larksuite/cli/releases/tag/v1.0.57
-[v1.0.56]: https://github.com/larksuite/cli/releases/tag/v1.0.56
-[v1.0.55]: https://github.com/larksuite/cli/releases/tag/v1.0.55
-[v1.0.54]: https://github.com/larksuite/cli/releases/tag/v1.0.54
-[v1.0.53]: https://github.com/larksuite/cli/releases/tag/v1.0.53
-[v1.0.52]: https://github.com/larksuite/cli/releases/tag/v1.0.52
-[v1.0.51]: https://github.com/larksuite/cli/releases/tag/v1.0.51
-[v1.0.50]: https://github.com/larksuite/cli/releases/tag/v1.0.50
-[v1.0.49]: https://github.com/larksuite/cli/releases/tag/v1.0.49
-[v1.0.48]: https://github.com/larksuite/cli/releases/tag/v1.0.48
-[v1.0.47]: https://github.com/larksuite/cli/releases/tag/v1.0.47
 [v1.0.46]: https://github.com/larksuite/cli/releases/tag/v1.0.46
 [v1.0.45]: https://github.com/larksuite/cli/releases/tag/v1.0.45
 [v1.0.44]: https://github.com/larksuite/cli/releases/tag/v1.0.44
--- a/52
+++ b/52
@@ -5,25 +5,10 @@ BINARY   := lark-cli
 MODULE   := github.com/larksuite/cli
 VERSION  := $(shell git describe --tags --always --dirty 2>/dev/null || echo dev)
 DATE     := $(shell date +%Y-%m-%d)
-NODE     ?= node
-QUALITY_GATE_CHANGED_FROM ?= $(shell bash scripts/resolve-changed-from.sh)
-QUALITY_GATE_CHANGED_FROM_RESOLVED = $(if $(strip $(QUALITY_GATE_CHANGED_FROM)),$(QUALITY_GATE_CHANGED_FROM),$(shell bash scripts/resolve-changed-from.sh))
-QUALITY_GATE_DIR ?= .tmp/quality-gate
-QUALITY_GATE_MANIFEST_OUT ?= $(QUALITY_GATE_DIR)/command-manifest.json
-QUALITY_GATE_COMMAND_INDEX_OUT ?= $(QUALITY_GATE_DIR)/command-index.json
-QUALITY_GATE_FACTS_OUT ?= $(QUALITY_GATE_DIR)/facts.json
-PUBLIC_CONTENT_METADATA ?= $(QUALITY_GATE_DIR)/public-content-metadata.json
 LDFLAGS  := -s -w -X $(MODULE)/internal/build.Version=$(VERSION) -X $(MODULE)/internal/build.Date=$(DATE)
 PREFIX   ?= /usr/local

-# The repository's Go 1.23 CI toolchain does not support -race on riscv64.
-# Prefer GOARCH passed to make (for example, `make GOARCH=riscv64 unit-test`)
-# over `go env GOARCH`, because command-line make variables are not visible to
-# $(shell ...).
-TEST_GOARCH := $(or $(GOARCH),$(shell go env GOARCH))
-RACE_FLAG := $(if $(filter riscv64,$(TEST_GOARCH)),,-race)
-
-.PHONY: all build vet fmt-check script-test test unit-test integration-test examples-build quality-gate install uninstall clean fetch_meta gitleaks
+.PHONY: all build vet fmt-check test unit-test integration-test examples-build install uninstall clean fetch_meta gitleaks

 all: test

@@ -47,15 +32,9 @@ fmt-check:
 		exit 1; \
 	fi

-script-test:
-	bash scripts/resolve-changed-from.test.sh
-	bash scripts/ci-workflow.test.sh
-	bash scripts/semantic-review-workflow.test.sh
-	$(NODE) --test scripts/semantic-review-verify-artifact.test.js scripts/pr-quality-summary.test.js scripts/semantic-review-publish.test.js scripts/ci-quality-summary-publish.test.js
-
 # ./extension/... keeps the public plugin SDK in the default test matrix.
 unit-test: fetch_meta
-	go test $(RACE_FLAG) -gcflags="all=-N -l" -count=1 \
+	go test -race -gcflags="all=-N -l" -count=1 \
 		./cmd/... ./internal/... ./shortcuts/... ./extension/...

 # examples-build keeps the shipped plugin-SDK examples compilable. If this
@@ -67,32 +46,7 @@ examples-build:
 integration-test: build
 	go test -v -count=1 ./tests/...

-test: vet fmt-check script-test unit-test examples-build integration-test
-
-quality-gate: build
-	mkdir -p $(QUALITY_GATE_DIR) $(dir $(QUALITY_GATE_FACTS_OUT)) $(dir $(PUBLIC_CONTENT_METADATA))
-	test -f $(PUBLIC_CONTENT_METADATA) || printf '{}\n' > $(PUBLIC_CONTENT_METADATA)
-	LARKSUITE_CLI_REMOTE_META=off \
-	LARKSUITE_CLI_NO_UPDATE_NOTIFIER=1 \
-	LARKSUITE_CLI_NO_SKILLS_NOTIFIER=1 \
-	go run ./internal/qualitygate/cmd/manifest-export \
-		--manifest-out $(QUALITY_GATE_MANIFEST_OUT) \
-		--command-index-out $(QUALITY_GATE_COMMAND_INDEX_OUT)
-	LARKSUITE_CLI_APP_ID=dry-run \
-	LARKSUITE_CLI_APP_SECRET=dry-run \
-	LARKSUITE_CLI_BRAND=feishu \
-	LARKSUITE_CLI_CONFIG_DIR=$${TMPDIR:-/tmp}/quality-gate-cli-config \
-	LARKSUITE_CLI_REMOTE_META=off \
-	LARKSUITE_CLI_NO_UPDATE_NOTIFIER=1 \
-	LARKSUITE_CLI_NO_SKILLS_NOTIFIER=1 \
-	go run ./internal/qualitygate/cmd/quality-gate check \
-		--repo . \
-		--cli-bin ./$(BINARY) \
-		--changed-from $(QUALITY_GATE_CHANGED_FROM_RESOLVED) \
-		--manifest $(QUALITY_GATE_MANIFEST_OUT) \
-		--command-index $(QUALITY_GATE_COMMAND_INDEX_OUT) \
-		--public-content-metadata $(PUBLIC_CONTENT_METADATA) \
-		--facts-out $(QUALITY_GATE_FACTS_OUT)
+test: vet fmt-check unit-test examples-build integration-test

 install: build
 	install -d $(PREFIX)/bin
--- a/README.md
+++ b/README.md
@@ -41,7 +41,7 @@ The official [Lark/Feishu](https://www.larksuite.com/) CLI tool, maintained by t
 | ✍️ Approval   | Query approval tasks, approve/reject/transfer tasks, cancel and CC instances                                                      |
 | 🎯 OKR        | Query, create, update OKRs; manage objective & key results, alignments, indicators and progress.                                  |
 | 📋 Project    | Meegle — manage work items, schedules, and data via the standalone [meegle-cli](https://github.com/larksuite/meegle-cli) (install separately) |
-| 🔗 Apps       | Create Spark/Miaoda apps, publish HTML/static sites, run cloud generation, and manage access scope                                 |
+| 🔗 Apps       | Develop, deploy HTML, web pages and applications                                                                                  |

 ## Installation & Quick Start

@@ -198,7 +198,7 @@ Prefixed with `+`, designed to be friendly for both humans and AI, with smart de
 ```bash
 lark-cli calendar +agenda
 lark-cli im +messages-send --chat-id "oc_xxx" --text "Hello"
-lark-cli docs +create --doc-format markdown --content $'<title>Weekly Report</title>\n# Progress\n- Completed feature X'
+lark-cli docs +create --api-version v2 --doc-format markdown --content $'<title>Weekly Report</title>\n# Progress\n- Completed feature X'
 ```

 Run `lark-cli <service> --help` to see all shortcut commands.
--- a/README.zh.md
+++ b/README.zh.md
@@ -41,7 +41,7 @@
 | ✍️ 审批     | 查询审批任务、同意/拒绝/转交审批任务、撤回与抄送审批实例              |
 | 🎯 OKR  | 查询、创建、更新 OKR，管理目标、关键结果、对齐、指标和进展记录        |
 | 📋 飞书项目 | 管理工作项、排期与数据 — 由独立的 [meegle-cli](https://github.com/larksuite/meegle-cli) 提供（需单独安装） |
-| 🔗 应用     | 创建妙搭（Spark/Miaoda）应用、发布 HTML/静态站点、云端生成迭代、管理可用范围                                                                                                   |
+| 🔗 应用     | 开发、部署 HTML、Web 页面和应用                                                                                                          |

 ## 安装与快速开始

@@ -199,7 +199,7 @@ CLI 提供三种粒度的调用方式，覆盖从快速操作到完全自定义
 ```bash
 lark-cli calendar +agenda
 lark-cli im +messages-send --chat-id "oc_xxx" --text "Hello"
-lark-cli docs +create --doc-format markdown --content $'<title>周报</title>\n# 本周进展\n- 完成了 X 功能'
+lark-cli docs +create --api-version v2 --doc-format markdown --content $'<title>周报</title>\n# 本周进展\n- 完成了 X 功能'
 ```

 运行 `lark-cli <service> --help` 查看所有快捷命令。
--- a/affordance/README.md
+++ b/affordance/README.md
@@ -1,49 +0,0 @@
-# Affordance
-
-Per-command usage guidance for the CLI, authored as one markdown file per domain
-(`<service>.md`). It is surfaced in `lark-cli <command> --help` and in the
-`schema` output, and read directly at runtime (lazy, cached) — there is no build
-step. Maintain these files alongside `skills/` and `shortcuts/`.
-
-## Format
-
-A small, fixed markdown subset; each file describes one domain:
-
-    # <domain>            optional `> skill: <name>` applies to every command below
-    ## <command>          the command as typed, minus `lark-cli <domain>`
-    <lead paragraph>      when to use this command
-    ### Avoid when        when not to use it / which command to use instead
-    ### Prerequisites     what you must have first (e.g. an id, and where it comes from)
-    ### Tips              gotchas and constraints
-    ### Examples          **description** lines, each followed by a fenced command
-    ### <other heading>   a custom section; flows through verbatim
-
-Reference another command with `[[command]]` — it renders as `command` in help.
-Under `Avoid when` it means "use that one instead"; under `Prerequisites`
-("… from [[command]]") it means "get the input there first".
-
-## Example
-
-    ## messages get
-    Fetch the full content of a single message by id.
-
-    ### Avoid when
-    - Reading several at once → use [[messages batch_get]]
-
-    ### Prerequisites
-    - message_id from [[messages list]]
-
-    ### Examples
-
-    **Fetch one message**
-    ```bash
-    lark-cli mail user_mailbox.messages get --message-id "<id>"
-    ```
-
-## Notes
-
- Write plain prose; the only convention is wrapping command references in `[[ ]]`.
- Keep it concise and high-signal — don't restate field/flag names, id types, or
-  anything the schema and flags already show; the agent infers the rest.
- Command-form headings resolve to method ids via the registry, so plural resource
-  names (`messages`) map to the singular method id (`message`) automatically.
--- a/affordance/contact.md
+++ b/affordance/contact.md
@@ -1,19 +0,0 @@
-# contact
-> skill: lark-contact
-
-## user_profiles batch_query
-Bulk-fetch personal status and signature for user ids you already have.
-
-### Avoid when
- Need more than status/signature (name, dept, email), or don't have the open_id yet → use [[+search-user]]
-
-### Tips
- Off by default — set include_personal_status / include_description to true under query_option
- ids in user_ids must match --user-id-type (default open_id)
-
-### Examples
-
-**Bulk-query status and signature**
-```bash
-lark-cli contact user_profiles batch_query --data '{"user_ids":["ou_3a8b****6a7b"],"query_option":{"include_personal_status":true,"include_description":true}}'
-```
--- a/cmd/api/api.go
+++ b/cmd/api/api.go
@@ -10,7 +10,6 @@ import (
 	"regexp"
 	"strings"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/client"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -67,21 +66,8 @@ func NewCmdApiWithContext(ctx context.Context, f *cmdutil.Factory, runF func(*AP

 	cmd := &cobra.Command{
 		Use:   "api <method> <path>",
-		Short: "Raw HTTP escape hatch — call any endpoint by path (fallback when no typed command exists)",
-		Long: `Raw HTTP escape hatch: send any Lark API request by HTTP method + path.
-
-Prefer the typed domain command when one exists — it validates parameters,
-shows the Risk level, gates destructive calls behind --yes, and carries usage
-guidance that this raw command does not. If a domain command covers your task
-(browse with ` + "`lark-cli <domain> --help`" + `), use it instead of this.
-
-Reach for ` + "`api`" + ` only for endpoints that have no typed command yet (e.g.
-newer/preview APIs), where you already have the HTTP path from the Lark docs.
-
-Examples:
-  lark-cli api GET /open-apis/calendar/v4/calendars
-  lark-cli api POST /open-apis/im/v1/messages --params '{"receive_id_type":"open_id"}' --data @body.json`,
-		Args: cobra.ExactArgs(2),
+		Short: "Generic Lark API requests",
+		Args:  cobra.ExactArgs(2),
 		RunE: func(cmd *cobra.Command, args []string) error {
 			opts.Method = strings.ToUpper(args[0])
 			opts.Path = args[1]
@@ -104,7 +90,6 @@ Examples:
 	cmd.Flags().IntVar(&opts.PageLimit, "page-limit", 10, "max pages to fetch with --page-all (0 = unlimited)")
 	cmd.Flags().IntVar(&opts.PageDelay, "page-delay", 200, "delay in ms between pages")
 	cmd.Flags().StringVar(&opts.Format, "format", "json", "output format: json|ndjson|table|csv")
-	cmd.Flags().Bool("json", false, "shorthand for --format json")
 	cmd.Flags().StringVarP(&opts.JqExpr, "jq", "q", "", "jq expression to filter JSON output")
 	cmd.Flags().BoolVar(&opts.DryRun, "dry-run", false, "print request without executing")
 	cmd.Flags().StringVar(&opts.File, "file", "", "file to upload as multipart/form-data ([field=]path, supports - for stdin)")
@@ -137,13 +122,7 @@ func buildAPIRequest(opts *APIOptions) (client.RawApiRequest, *cmdutil.FileUploa

 	// stdin conflict: --params and --data cannot both read from stdin, regardless of --file.
 	if opts.Params == "-" && opts.Data == "-" {
-		return client.RawApiRequest{}, nil, errs.NewValidationError(errs.SubtypeInvalidArgument,
-			"--params and --data cannot both read from stdin (-)").
-			WithHint("pass at most one flag as '-'; give the other inline JSON or @file").
-			WithParams(
-				errs.InvalidParam{Name: "--params", Reason: "reads from stdin (-)"},
-				errs.InvalidParam{Name: "--data", Reason: "reads from stdin (-)"},
-			)
+		return client.RawApiRequest{}, nil, output.ErrValidation("--params and --data cannot both read from stdin (-)")
 	}

 	params, err := cmdutil.ParseJSONMap(opts.Params, "--params", stdin, fileIO)
@@ -173,10 +152,7 @@ func buildAPIRequest(opts *APIOptions) (client.RawApiRequest, *cmdutil.FileUploa
 				return client.RawApiRequest{}, nil, err
 			}
 			if _, ok := dataFields.(map[string]any); !ok {
-				return client.RawApiRequest{}, nil, errs.NewValidationError(errs.SubtypeInvalidArgument,
-					"--data must be a JSON object when used with --file").
-					WithHint(`with --file, --data carries multipart form fields, e.g. --data '{"image_type":"message"}'`).
-					WithParam("--data")
+				return client.RawApiRequest{}, nil, output.ErrValidation("--data must be a JSON object when used with --file")
 			}
 		}

@@ -219,13 +195,7 @@ func apiRun(opts *APIOptions) error {
 	}

 	if opts.PageAll && opts.Output != "" {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument,
-			"--output and --page-all are mutually exclusive").
-			WithHint("drop --page-all to save a binary response, or drop --output to paginate JSON").
-			WithParams(
-				errs.InvalidParam{Name: "--output", Reason: "conflicts with --page-all"},
-				errs.InvalidParam{Name: "--page-all", Reason: "conflicts with --output"},
-			)
+		return output.ErrValidation("--output and --page-all are mutually exclusive")
 	}
 	if err := output.ValidateJqFlags(opts.JqExpr, opts.Output, opts.Format); err != nil {
 		return err
@@ -262,7 +232,7 @@ func apiRun(opts *APIOptions) error {
 	}

 	if opts.PageAll {
-		return apiPaginate(opts.Ctx, ac, request, format, opts.JqExpr, out, f.IOStreams.ErrOut, opts.Cmd.CommandPath(),
+		return apiPaginate(opts.Ctx, ac, request, format, opts.JqExpr, out, f.IOStreams.ErrOut,
 			client.PaginationOptions{PageLimit: opts.PageLimit, PageDelay: opts.PageDelay})
 	}

@@ -272,7 +242,7 @@ func apiRun(opts *APIOptions) error {
 		// pass on *output.ExitError values. Typed *errs.* errors that flow
 		// through here keep their canonical message / hint from BuildAPIError;
 		// MarkRaw is a no-op on those (it only flips a flag on *ExitError).
-		return errs.MarkRaw(err)
+		return output.MarkRaw(err)
 	}
 	err = client.HandleResponse(resp, client.ResponseOptions{
 		OutputPath:  opts.Output,
@@ -292,7 +262,7 @@ func apiRun(opts *APIOptions) error {
 	// MarkRaw: see comment above on the DoAPI path. Skips legacy
 	// *ExitError enrichment; typed errors flow through unchanged.
 	if err != nil {
-		return errs.MarkRaw(err)
+		return output.MarkRaw(err)
 	}
 	return nil
 }
@@ -301,76 +271,46 @@ func apiDryRun(f *cmdutil.Factory, request client.RawApiRequest, config *core.Cl
 	return cmdutil.PrintDryRun(f.IOStreams.Out, request, config, format)
 }

-func apiPaginate(ctx context.Context, ac *client.APIClient, request client.RawApiRequest, format output.Format, jqExpr string, out, errOut io.Writer, commandPath string, pagOpts client.PaginationOptions) error {
+func apiPaginate(ctx context.Context, ac *client.APIClient, request client.RawApiRequest, format output.Format, jqExpr string, out, errOut io.Writer, pagOpts client.PaginationOptions) error {
 	if pagOpts.Identity == "" {
 		pagOpts.Identity = request.As
 	}
 	// When jq is set, always aggregate all pages then filter.
 	if jqExpr != "" {
-		result, err := ac.PaginateAll(ctx, request, pagOpts)
-		if err != nil {
-			return errs.MarkRaw(err)
+		if err := client.PaginateWithJq(ctx, ac, request, jqExpr, out, pagOpts, ac.CheckResponse); err != nil {
+			return output.MarkRaw(err)
 		}
-		if apiErr := ac.CheckResponse(result, pagOpts.Identity); apiErr != nil {
-			output.FormatValue(out, result, output.FormatJSON)
-			return errs.MarkRaw(apiErr)
-		}
-		return output.WriteSuccessEnvelope(output.SuccessEnvelopeData(result), output.SuccessEnvelopeOptions{
-			CommandPath: commandPath,
-			Identity:    string(pagOpts.Identity),
-			JqExpr:      jqExpr,
-			Out:         out,
-			ErrOut:      errOut,
-		})
+		return nil
 	}

 	switch format {
 	case output.FormatNDJSON, output.FormatTable, output.FormatCSV:
 		pf := output.NewPaginatedFormatter(out, format)
-		result, hasItems, err := ac.StreamPages(ctx, request, func(items []interface{}) error {
-			// Streaming formats intentionally emit each page after that page has
-			// passed safety scanning. A later page may still fail, so callers
-			// must use the exit code to distinguish complete vs partial output.
-			scanResult := output.ScanForSafety(commandPath, items, errOut)
-			if scanResult.Blocked {
-				return scanResult.BlockErr
-			}
-			if scanResult.Alert != nil {
-				output.WriteAlertWarning(errOut, scanResult.Alert)
-			}
+		result, hasItems, err := ac.StreamPages(ctx, request, func(items []interface{}) {
 			pf.FormatPage(items)
-			return nil
 		}, pagOpts)
 		if err != nil {
-			return errs.MarkRaw(err)
+			return output.MarkRaw(err)
 		}
 		if apiErr := ac.CheckResponse(result, pagOpts.Identity); apiErr != nil {
-			return errs.MarkRaw(apiErr)
+			output.FormatValue(out, result, output.FormatJSON)
+			return output.MarkRaw(apiErr)
 		}
 		if !hasItems {
 			fmt.Fprintf(errOut, "warning: this API does not return a list, format %q is not supported, falling back to json\n", format)
-			return output.WriteSuccessEnvelope(output.SuccessEnvelopeData(result), output.SuccessEnvelopeOptions{
-				CommandPath: commandPath,
-				Identity:    string(pagOpts.Identity),
-				Out:         out,
-				ErrOut:      errOut,
-			})
+			output.FormatValue(out, result, output.FormatJSON)
 		}
 		return nil
 	default:
 		result, err := ac.PaginateAll(ctx, request, pagOpts)
 		if err != nil {
-			return errs.MarkRaw(err)
+			return output.MarkRaw(err)
 		}
 		if apiErr := ac.CheckResponse(result, pagOpts.Identity); apiErr != nil {
 			output.FormatValue(out, result, output.FormatJSON)
-			return errs.MarkRaw(apiErr)
+			return output.MarkRaw(apiErr)
 		}
-		return output.WriteSuccessEnvelope(output.SuccessEnvelopeData(result), output.SuccessEnvelopeOptions{
-			CommandPath: commandPath,
-			Identity:    string(pagOpts.Identity),
-			Out:         out,
-			ErrOut:      errOut,
-		})
+		output.FormatValue(out, result, format)
+		return nil
 	}
 }
--- a/cmd/api/api_test.go
+++ b/cmd/api/api_test.go
@@ -4,8 +4,6 @@
 package api

 import (
-	"context"
-	"encoding/json"
 	"errors"
 	"os"
 	"sort"
@@ -13,7 +11,6 @@ import (
 	"testing"

 	"github.com/larksuite/cli/errs"
-	extcs "github.com/larksuite/cli/extension/contentsafety"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/httpmock"
@@ -69,24 +66,6 @@ func TestApiCmd_DryRun(t *testing.T) {
 	}
 }

-// Regression: --params null parses to a nil map; writing page_size onto it must
-// not panic. Symmetric to the typed-flag overlay path in cmd/service — both
-// write into the map ParseJSONMap returns.
-func TestApiCmd_NullParamsWithPageSize(t *testing.T) {
-	f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
-	})
-
-	cmd := NewCmdApi(f, nil)
-	cmd.SetArgs([]string{"GET", "/open-apis/test", "--params", "null", "--page-size", "50", "--as", "bot", "--dry-run"})
-	if err := cmd.Execute(); err != nil {
-		t.Fatalf("--params null with --page-size should not error, got: %v", err)
-	}
-	if out := stdout.String(); !strings.Contains(out, "page_size") {
-		t.Errorf("expected page_size applied over null --params, got:\n%s", out)
-	}
-}
-
 func TestApiCmd_BotMode(t *testing.T) {
 	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
 		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
@@ -104,19 +83,8 @@ func TestApiCmd_BotMode(t *testing.T) {
 	if err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	var got map[string]interface{}
-	if err := json.Unmarshal(stdout.Bytes(), &got); err != nil {
-		t.Fatalf("invalid JSON output: %v\n%s", err, stdout.String())
-	}
-	if got["ok"] != true || got["identity"] != "bot" {
-		t.Fatalf("unexpected envelope: %#v", got)
-	}
-	if _, hasCode := got["code"]; hasCode {
-		t.Fatalf("success envelope leaked outer code: %s", stdout.String())
-	}
-	data, ok := got["data"].(map[string]interface{})
-	if !ok || data["result"] != "success" {
-		t.Fatalf("data = %#v, want result=success", got["data"])
+	if !strings.Contains(stdout.String(), "success") {
+		t.Error("expected 'success' in output")
 	}
 }

@@ -342,16 +310,8 @@ func TestApiCmd_PageAll_NonBatchAPI_FallbackToJSON(t *testing.T) {
 		t.Error("expected 'falling back to json' in stderr")
 	}
 	// Should output JSON result to stdout
-	var got map[string]interface{}
-	if err := json.Unmarshal(stdout.Bytes(), &got); err != nil {
-		t.Fatalf("invalid JSON output: %v\n%s", err, stdout.String())
-	}
-	data, ok := got["data"].(map[string]interface{})
-	if got["ok"] != true || got["identity"] != "bot" || !ok || data["user_id"] != "u123" {
-		t.Fatalf("unexpected fallback envelope: %#v", got)
-	}
-	if _, hasCode := got["code"]; hasCode {
-		t.Fatalf("fallback success envelope leaked outer code: %s", stdout.String())
+	if !strings.Contains(stdout.String(), "u123") {
+		t.Error("expected user_id in JSON output")
 	}
 }

@@ -364,7 +324,7 @@ func TestApiCmd_PageAll_NonBatchAPI_ErrorStillOutputsJSON(t *testing.T) {
 	reg.Register(&httpmock.Stub{
 		URL: "/open-apis/im/v1/chats/oc_xxx/announcement",
 		Body: map[string]interface{}{
-			"code": 230027, "msg": "user not authorized",
+			"code": 230001, "msg": "no permission",
 		},
 	})

@@ -376,20 +336,12 @@ func TestApiCmd_PageAll_NonBatchAPI_ErrorStillOutputsJSON(t *testing.T) {
 		t.Fatal("expected an error for non-zero code")
 	}
 	// Should still output the response body so user can see the error details
-	if !strings.Contains(stdout.String(), "230027") {
+	if !strings.Contains(stdout.String(), "230001") {
 		t.Errorf("expected error response in stdout, got: %s", stdout.String())
 	}
-	if !strings.Contains(stdout.String(), "user not authorized") {
+	if !strings.Contains(stdout.String(), "no permission") {
 		t.Errorf("expected error message in stdout, got: %s", stdout.String())
 	}
-	if strings.Contains(stdout.String(), `"ok": true`) || strings.Contains(stdout.String(), `"ok":true`) {
-		t.Fatalf("unexpected success envelope on error path: %s", stdout.String())
-	}
-	requireProblem(t, err, errs.CategoryAuthorization, errs.SubtypeUserUnauthorized, 230027)
-	var permErr *errs.PermissionError
-	if !errors.As(err, &permErr) {
-		t.Fatalf("expected PermissionError, got %T: %v", err, err)
-	}
 }

 func TestApiCmd_PageAll_BatchAPI_StreamsItems(t *testing.T) {
@@ -425,274 +377,6 @@ func TestApiCmd_PageAll_BatchAPI_StreamsItems(t *testing.T) {
 	}
 }

-func TestApiCmd_PageAll_StreamBusinessErrorDoesNotDumpJSON(t *testing.T) {
-	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app-pageall-stream-err", AppSecret: "test-secret-pageall-stream-err", Brand: core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":      []interface{}{map[string]interface{}{"id": "safe-page"}},
-				"has_more":   true,
-				"page_token": "next",
-			},
-		},
-	})
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 230027, "msg": "user not authorized",
-		},
-	})
-
-	cmd := NewCmdApi(f, nil)
-	cmd.SetArgs([]string{"GET", "/open-apis/contact/v3/users", "--as", "bot", "--page-all", "--format", "ndjson"})
-	err := cmd.Execute()
-	if err == nil {
-		t.Fatal("expected error for non-zero code on later page")
-	}
-	requireProblem(t, err, errs.CategoryAuthorization, errs.SubtypeUserUnauthorized, 230027)
-	out := stdout.String()
-	if !strings.Contains(out, "safe-page") {
-		t.Fatalf("expected earlier successful page to remain streamed, got: %s", out)
-	}
-	if strings.Contains(out, "230027") || strings.Contains(out, "user not authorized") {
-		t.Fatalf("streaming stdout should not contain raw error JSON, got: %s", out)
-	}
-	if strings.Contains(out, "\n  \"code\"") {
-		t.Fatalf("streaming stdout should not contain indented JSON error dump, got: %s", out)
-	}
-}
-
-func TestApiCmd_PageAll_BatchAPI_DefaultJSONEnvelope(t *testing.T) {
-	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app-pageall-json", AppSecret: "test-secret-pageall-json", Brand: core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":    []interface{}{map[string]interface{}{"id": "1"}},
-				"has_more": false,
-			},
-		},
-	})
-
-	cmd := NewCmdApi(f, nil)
-	cmd.SetArgs([]string{"GET", "/open-apis/contact/v3/users", "--as", "bot", "--page-all"})
-	if err := cmd.Execute(); err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-
-	var got map[string]interface{}
-	if err := json.Unmarshal(stdout.Bytes(), &got); err != nil {
-		t.Fatalf("invalid JSON output: %v\n%s", err, stdout.String())
-	}
-	data, ok := got["data"].(map[string]interface{})
-	if got["ok"] != true || got["identity"] != "bot" || !ok {
-		t.Fatalf("unexpected envelope: %#v", got)
-	}
-	if _, hasCode := got["code"]; hasCode {
-		t.Fatalf("success envelope leaked outer code: %s", stdout.String())
-	}
-	items, ok := data["items"].([]interface{})
-	if !ok || len(items) != 1 {
-		t.Fatalf("data.items = %#v, want one item", data["items"])
-	}
-}
-
-type apiContentSafetyProvider struct {
-	called bool
-	path   string
-	data   interface{}
-	match  string
-}
-
-func (p *apiContentSafetyProvider) Name() string { return "api-test" }
-
-func (p *apiContentSafetyProvider) Scan(_ context.Context, req extcs.ScanRequest) (*extcs.Alert, error) {
-	p.called = true
-	p.path = req.Path
-	p.data = req.Data
-	if p.match != "" {
-		b, _ := json.Marshal(req.Data)
-		if !strings.Contains(string(b), p.match) {
-			return nil, nil
-		}
-	}
-	return &extcs.Alert{Provider: "api-test", MatchedRules: []string{"pagination"}}, nil
-}
-
-func TestApiCmd_PageAll_DefaultJSONRunsContentSafety(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONTENT_SAFETY_MODE", "warn")
-	provider := &apiContentSafetyProvider{}
-	extcs.Register(provider)
-	t.Cleanup(func() { extcs.Register(nil) })
-
-	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app-pageall-safety", AppSecret: "test-secret-pageall-safety", Brand: core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":    []interface{}{map[string]interface{}{"id": "1"}},
-				"has_more": false,
-			},
-		},
-	})
-
-	root := &cobra.Command{Use: "lark-cli"}
-	root.AddCommand(NewCmdApi(f, nil))
-	root.SetArgs([]string{"api", "GET", "/open-apis/contact/v3/users", "--as", "bot", "--page-all"})
-	if err := root.Execute(); err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if !provider.called {
-		t.Fatal("expected content safety provider to scan paginated output")
-	}
-	if provider.path != "api" {
-		t.Fatalf("scan path = %q, want api", provider.path)
-	}
-	data, ok := provider.data.(map[string]interface{})
-	if !ok {
-		t.Fatalf("scanned data type = %T, want map", provider.data)
-	}
-	if _, hasCode := data["code"]; hasCode {
-		t.Fatalf("scanned data should be business data only, got %#v", data)
-	}
-
-	var got map[string]interface{}
-	if err := json.Unmarshal(stdout.Bytes(), &got); err != nil {
-		t.Fatalf("invalid JSON output: %v\n%s", err, stdout.String())
-	}
-	alert, ok := got["_content_safety_alert"].(map[string]interface{})
-	if !ok || alert["provider"] != "api-test" {
-		t.Fatalf("missing content safety alert in envelope: %#v", got)
-	}
-}
-
-func TestApiCmd_PageAll_StreamFormatRunsContentSafety(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONTENT_SAFETY_MODE", "warn")
-	provider := &apiContentSafetyProvider{}
-	extcs.Register(provider)
-	t.Cleanup(func() { extcs.Register(nil) })
-
-	f, stdout, stderr, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app-pageall-stream-safety", AppSecret: "test-secret-pageall-stream-safety", Brand: core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":    []interface{}{map[string]interface{}{"id": "1"}},
-				"has_more": false,
-			},
-		},
-	})
-
-	root := &cobra.Command{Use: "lark-cli"}
-	root.AddCommand(NewCmdApi(f, nil))
-	root.SetArgs([]string{"api", "GET", "/open-apis/contact/v3/users", "--as", "bot", "--page-all", "--format", "ndjson"})
-	if err := root.Execute(); err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if !provider.called {
-		t.Fatal("expected content safety provider to scan streamed paginated output")
-	}
-	if provider.path != "api" {
-		t.Fatalf("scan path = %q, want api", provider.path)
-	}
-	items, ok := provider.data.([]interface{})
-	if !ok || len(items) != 1 {
-		t.Fatalf("scanned data = %#v, want one streamed item", provider.data)
-	}
-	if !strings.Contains(stderr.String(), "warning: content safety alert from api-test") {
-		t.Fatalf("expected content safety warning on stderr, got: %s", stderr.String())
-	}
-	if !strings.Contains(stdout.String(), `"id":"1"`) {
-		t.Fatalf("expected streamed ndjson output, got: %s", stdout.String())
-	}
-}
-
-func TestApiCmd_PageAll_StreamFormatBlockSkipsBlockedPage(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONTENT_SAFETY_MODE", "block")
-	provider := &apiContentSafetyProvider{match: "blocked"}
-	extcs.Register(provider)
-	t.Cleanup(func() { extcs.Register(nil) })
-
-	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app-pageall-stream-block", AppSecret: "test-secret-pageall-stream-block", Brand: core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":      []interface{}{map[string]interface{}{"id": "safe-page"}},
-				"has_more":   true,
-				"page_token": "next",
-			},
-		},
-	})
-	reg.Register(&httpmock.Stub{
-		URL: "/open-apis/contact/v3/users",
-		Body: map[string]interface{}{
-			"code": 0, "msg": "ok",
-			"data": map[string]interface{}{
-				"items":    []interface{}{map[string]interface{}{"id": "blocked-page"}},
-				"has_more": false,
-			},
-		},
-	})
-
-	root := &cobra.Command{Use: "lark-cli"}
-	root.AddCommand(NewCmdApi(f, nil))
-	root.SetArgs([]string{"api", "GET", "/open-apis/contact/v3/users", "--as", "bot", "--page-all", "--format", "ndjson"})
-	err := root.Execute()
-	if err == nil {
-		t.Fatal("expected content safety block error")
-	}
-	var safetyErr *errs.ContentSafetyError
-	if !errors.As(err, &safetyErr) {
-		t.Fatalf("expected ContentSafetyError, got %T: %v", err, err)
-	}
-	if safetyErr.Category != errs.CategoryPolicy || safetyErr.Subtype != errs.SubtypeContentSafety {
-		t.Fatalf("problem = %s/%s, want %s/%s", safetyErr.Category, safetyErr.Subtype, errs.CategoryPolicy, errs.SubtypeContentSafety)
-	}
-	if len(safetyErr.Rules) != 1 || safetyErr.Rules[0] != "pagination" {
-		t.Fatalf("rules = %v, want [pagination]", safetyErr.Rules)
-	}
-	out := stdout.String()
-	if !strings.Contains(out, "safe-page") {
-		t.Fatalf("expected earlier safe page to remain streamed, got: %s", out)
-	}
-	if strings.Contains(out, "blocked-page") {
-		t.Fatalf("blocked page was written before safety block: %s", out)
-	}
-}
-
-func requireProblem(t *testing.T, err error, category errs.Category, subtype errs.Subtype, code int) {
-	t.Helper()
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed error, got %T: %v", err, err)
-	}
-	if p.Category != category || p.Subtype != subtype || p.Code != code {
-		t.Fatalf("problem = %s/%s/%d, want %s/%s/%d", p.Category, p.Subtype, p.Code, category, subtype, code)
-	}
-}
-
 func TestNormalisePath_StripsQueryAndFragment(t *testing.T) {
 	for _, tt := range []struct {
 		name string
@@ -1034,23 +718,3 @@ func TestApiCmd_PermissionError_DerivesFirstClassFields(t *testing.T) {
 		t.Errorf("LogID = %q, want %q", pe.LogID, "20260527-test-log")
 	}
 }
-
-func TestApiCmd_JsonFlag_Accepted(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
-	})
-
-	var gotOpts *APIOptions
-	cmd := NewCmdApi(f, func(opts *APIOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"GET", "/open-apis/test", "--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("--json should be accepted without error, got: %v", err)
-	}
-	if gotOpts.Method != "GET" {
-		t.Errorf("expected method GET, got %s", gotOpts.Method)
-	}
-}
--- a/cmd/auth/auth_test.go
+++ b/cmd/auth/auth_test.go
@@ -91,29 +91,6 @@ func TestAuthCheckCmd_FlagParsing(t *testing.T) {
 	}
 }

-func TestAuthCheckCmd_AcceptsJSONFlag(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
-	})
-
-	var gotOpts *CheckOptions
-	cmd := NewCmdAuthCheck(f, func(opts *CheckOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"--scope", "calendar:calendar:read", "--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if gotOpts == nil {
-		t.Fatal("expected opts to be set")
-	}
-	if !gotOpts.JSON {
-		t.Error("expected JSON=true")
-	}
-}
-
 func TestAuthLogoutCmd_FlagParsing(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)

@@ -132,27 +109,6 @@ func TestAuthLogoutCmd_FlagParsing(t *testing.T) {
 	}
 }

-func TestAuthLogoutCmd_AcceptsJSONFlag(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, nil)
-
-	var gotOpts *LogoutOptions
-	cmd := NewCmdAuthLogout(f, func(opts *LogoutOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if gotOpts == nil {
-		t.Fatal("expected opts to be set")
-	}
-	if !gotOpts.JSON {
-		t.Error("expected JSON=true")
-	}
-}
-
 func TestAuthListCmd_FlagParsing(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)

@@ -170,27 +126,6 @@ func TestAuthListCmd_FlagParsing(t *testing.T) {
 	}
 }

-func TestAuthListCmd_AcceptsJSONFlag(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, nil)
-
-	var gotOpts *ListOptions
-	cmd := NewCmdAuthList(f, func(opts *ListOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if gotOpts == nil {
-		t.Error("expected opts to be set")
-	}
-	if !gotOpts.JSON {
-		t.Error("expected JSON=true")
-	}
-}
-
 func TestAuthStatusCmd_FlagParsing(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
 		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
@@ -210,29 +145,6 @@ func TestAuthStatusCmd_FlagParsing(t *testing.T) {
 	}
 }

-func TestAuthStatusCmd_AcceptsJSONFlag(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
-	})
-
-	var gotOpts *StatusOptions
-	cmd := NewCmdAuthStatus(f, func(opts *StatusOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if gotOpts == nil {
-		t.Error("expected opts to be set")
-	}
-	if !gotOpts.JSON {
-		t.Error("expected JSON=true")
-	}
-}
-
 func TestAuthStatusCmd_VerifyFlag(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
 		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
@@ -355,32 +267,6 @@ func TestAuthScopesCmd_FlagParsing(t *testing.T) {
 	}
 }

-func TestAuthScopesCmd_JSONFlagForcesJSONFormat(t *testing.T) {
-	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu,
-	})
-
-	var gotOpts *ScopesOptions
-	cmd := NewCmdAuthScopes(f, func(opts *ScopesOptions) error {
-		gotOpts = opts
-		return nil
-	})
-	cmd.SetArgs([]string{"--format", "pretty", "--json"})
-	err := cmd.Execute()
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if gotOpts == nil {
-		t.Fatal("expected opts to be set")
-	}
-	if !gotOpts.JSON {
-		t.Error("expected JSON=true")
-	}
-	if gotOpts.Format != "json" {
-		t.Errorf("expected format json, got %s", gotOpts.Format)
-	}
-}
-
 func TestAuthScopesRun_UsesTenantAccessTokenFromCredentialProvider(t *testing.T) {
 	f, _, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
 		AppID: "test-app", AppSecret: "", Brand: core.BrandFeishu,
--- a/cmd/auth/check.go
+++ b/cmd/auth/check.go
@@ -19,7 +19,6 @@ import (
 type CheckOptions struct {
 	Factory *cmdutil.Factory
 	Scope   string
-	JSON    bool
 }

 // NewCmdAuthCheck creates the auth check subcommand.
@@ -38,7 +37,6 @@ func NewCmdAuthCheck(f *cmdutil.Factory, runF func(*CheckOptions) error) *cobra.
 	}

 	cmd.Flags().StringVar(&opts.Scope, "scope", "", "scopes to check (space-separated)")
-	cmd.Flags().BoolVar(&opts.JSON, "json", false, "structured JSON output")
 	cmd.MarkFlagRequired("scope")
 	cmdutil.SetRisk(cmd, "read")

--- a/cmd/auth/check_test.go
+++ b/cmd/auth/check_test.go
@@ -33,9 +33,12 @@ func TestAuthCheckRun_NotLoggedIn_ExitOneWithStdoutOnly(t *testing.T) {
 	if got := output.ExitCodeOf(err); got != 1 {
 		t.Errorf("exit code = %d, want 1 (predicate 'missing' signal)", got)
 	}
-	var bare *output.BareError
+	var bare *output.ExitError
 	if !errors.As(err, &bare) {
-		t.Fatalf("expected *output.BareError (ErrBare), got %T: %v", err, err)
+		t.Fatalf("expected *output.ExitError (ErrBare), got %T: %v", err, err)
+	}
+	if bare.Detail != nil {
+		t.Errorf("ErrBare must carry no Detail (no envelope), got %+v", bare.Detail)
 	}

 	if stderr.Len() != 0 {
--- a/cmd/auth/list.go
+++ b/cmd/auth/list.go
@@ -9,7 +9,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	larkauth "github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -19,7 +18,6 @@ import (
 // ListOptions holds all inputs for auth list.
 type ListOptions struct {
 	Factory *cmdutil.Factory
-	JSON    bool
 }

 // NewCmdAuthList creates the auth list subcommand.
@@ -36,7 +34,6 @@ func NewCmdAuthList(f *cmdutil.Factory, runF func(*ListOptions) error) *cobra.Co
 			return authListRun(opts)
 		},
 	}
-	cmd.Flags().BoolVar(&opts.JSON, "json", false, "structured JSON output")
 	cmdutil.SetRisk(cmd, "read")

 	return cmd
@@ -47,20 +44,12 @@ func authListRun(opts *ListOptions) error {

 	multi, _ := core.LoadMultiAppConfig()
 	if multi == nil || len(multi.Apps) == 0 {
-		if opts.JSON {
-			output.PrintJson(f.IOStreams.Out, map[string]interface{}{
-				"ok":     true,
-				"users":  []map[string]interface{}{},
-				"reason": "not_configured",
-			})
-			return nil
-		}
 		// auth list is a read-only probe; the "configured but no users"
 		// branch below already returns exit 0 with a stderr hint, so we
 		// keep the same contract here. We still want the hint to be
 		// workspace-aware, so we pull the message+hint out of
 		// NotConfiguredError() instead of hard-coding it.
-		var cfgErr *errs.ConfigError
+		var cfgErr *core.ConfigError
 		if errors.As(core.NotConfiguredError(), &cfgErr) {
 			fmt.Fprintln(f.IOStreams.ErrOut, cfgErr.Message)
 			if cfgErr.Hint != "" {
@@ -72,14 +61,6 @@ func authListRun(opts *ListOptions) error {

 	app := multi.CurrentAppConfig(f.Invocation.Profile)
 	if app == nil || len(app.Users) == 0 {
-		if opts.JSON {
-			output.PrintJson(f.IOStreams.Out, map[string]interface{}{
-				"ok":     true,
-				"users":  []map[string]interface{}{},
-				"reason": "not_logged_in",
-			})
-			return nil
-		}
 		fmt.Fprintln(f.IOStreams.ErrOut, "No logged-in users. Run `lark-cli auth login` to log in.")
 		return nil
 	}
--- a/cmd/auth/list_test.go
+++ b/cmd/auth/list_test.go
@@ -4,7 +4,6 @@
 package auth

 import (
-	"encoding/json"
 	"strings"
 	"testing"

@@ -35,33 +34,6 @@ func TestAuthListRun_NotConfigured_ReturnsExitZero(t *testing.T) {
 	}
 }

-func TestAuthListRun_JSONMode_NotConfigured_WritesStdoutOnly(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authListRun(&ListOptions{Factory: f, JSON: true}); err != nil {
-		t.Fatalf("auth list should succeed when not configured (exit 0); got: %v", err)
-	}
-
-	var payload map[string]any
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("stdout must be valid JSON: %v\nstdout=%s", err, stdout.String())
-	}
-	if payload["ok"] != true {
-		t.Errorf("stdout.ok = %v, want true", payload["ok"])
-	}
-	users, ok := payload["users"].([]any)
-	if !ok || len(users) != 0 {
-		t.Errorf("stdout.users = %v, want empty array", payload["users"])
-	}
-	if payload["reason"] != "not_configured" {
-		t.Errorf("stdout.reason = %v, want not_configured", payload["reason"])
-	}
-	if stderr.Len() != 0 {
-		t.Errorf("stderr must stay empty in JSON mode, got:\n%s", stderr.String())
-	}
-}
-
 // TestAuthListRun_NotConfigured_AgentWorkspace_RoutesToBindHelp covers the
 // reason this hint exists workspace-aware in the first place: an AI agent
 // in OpenClaw / Hermes that probes auth list before binding gets routed to
@@ -85,48 +57,3 @@ func TestAuthListRun_NotConfigured_AgentWorkspace_RoutesToBindHelp(t *testing.T)
 		t.Errorf("agent hint must not mention config init: %s", out)
 	}
 }
-
-func TestAuthListRun_JSONMode_NoLoggedInUsers_WritesStdoutOnly(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	writeLogoutConfig(t, nil)
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authListRun(&ListOptions{Factory: f, JSON: true}); err != nil {
-		t.Fatalf("auth list should succeed when no users exist (exit 0); got: %v", err)
-	}
-
-	var payload map[string]any
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("stdout must be valid JSON: %v\nstdout=%s", err, stdout.String())
-	}
-	if payload["ok"] != true {
-		t.Errorf("stdout.ok = %v, want true", payload["ok"])
-	}
-	users, ok := payload["users"].([]any)
-	if !ok || len(users) != 0 {
-		t.Errorf("stdout.users = %v, want empty array", payload["users"])
-	}
-	if payload["reason"] != "not_logged_in" {
-		t.Errorf("stdout.reason = %v, want not_logged_in", payload["reason"])
-	}
-	if stderr.Len() != 0 {
-		t.Errorf("stderr must stay empty in JSON mode, got:\n%s", stderr.String())
-	}
-}
-
-func TestAuthListRun_DefaultMode_NoLoggedInUsers_KeepsTextOutput(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	writeLogoutConfig(t, nil)
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authListRun(&ListOptions{Factory: f}); err != nil {
-		t.Fatalf("auth list should succeed when no users exist (exit 0); got: %v", err)
-	}
-
-	if stdout.Len() != 0 {
-		t.Errorf("stdout must stay empty in default mode, got:\n%s", stdout.String())
-	}
-	if !strings.Contains(stderr.String(), "No logged-in users") {
-		t.Errorf("stderr = %q, want no-users hint", stderr.String())
-	}
-}
--- a/cmd/auth/login.go
+++ b/cmd/auth/login.go
@@ -296,11 +296,10 @@ func authLoginRun(opts *LoginOptions) error {
 	}

 	// Step 2: Show user code and verification URL.
-	// JSON mode embeds AgentTimeoutHint as a structured field so agents that
-	// capture stdout into a JSON parser see it without stream-mixing surprises.
-	// Text mode prints the hint to stderr only when running under a non-TTY
-	// (i.e. piped / agent harness), since humans reading a terminal don't need
-	// the agent-oriented instructions.
+	// Both branches surface AgentTimeoutHint, but on different channels:
+	// JSON mode embeds it as a structured field (so an agent that captures
+	// stdout into a JSON parser sees it without stream-mixing surprises),
+	// text mode prints to stderr (alongside the URL prompt).
 	if opts.JSON {
 		data := map[string]interface{}{
 			"event":                     "device_authorization",
@@ -318,9 +317,7 @@ func authLoginRun(opts *LoginOptions) error {
 	} else {
 		fmt.Fprintf(f.IOStreams.ErrOut, msg.OpenURL)
 		fmt.Fprintf(f.IOStreams.ErrOut, "  %s\n\n", authResp.VerificationUriComplete)
-		if f.IOStreams != nil && !f.IOStreams.IsTerminal {
-			fmt.Fprintln(f.IOStreams.ErrOut, msg.AgentTimeoutHint)
-		}
+		fmt.Fprintln(f.IOStreams.ErrOut, msg.AgentTimeoutHint)
 	}

 	// Step 3: Poll for token
@@ -407,11 +404,10 @@ func authLoginPollDeviceCode(opts *LoginOptions, config *core.CliConfig, msg *lo
 			fmt.Fprintf(f.IOStreams.ErrOut, "[lark-cli] [WARN] auth login: failed to remove cached requested scopes: %v\n", err)
 		}
 	}
-	// Skip the stderr hint in JSON mode (the --no-wait call that issued
-	// the device_code already surfaced it as a JSON field), and also skip it
-	// when running on an interactive terminal — the agent-oriented
-	// instructions only matter for piped / harness environments.
-	if !opts.JSON && f.IOStreams != nil && !f.IOStreams.IsTerminal {
+	// Skip the stderr hint in JSON mode — the --no-wait call that issued the
+	// device_code already returned the hint as a JSON field, and writing
+	// text to stderr would pollute consumers that combine streams via 2>&1.
+	if !opts.JSON {
 		fmt.Fprintln(f.IOStreams.ErrOut, msg.AgentTimeoutHint)
 	}
 	log(msg.WaitingAuth)
@@ -531,10 +527,10 @@ func collectScopesForDomains(domains []string, identity string, brand core.LarkB

 	// 3. Shortcut scopes matching by Service (only include shortcuts supporting the identity)
 	for _, sc := range shortcuts.AllShortcuts() {
-		if !shortcuts.IsShortcutServiceAvailable(sc.Service, brand) {
+		if !shortcuts.IsShortcutServiceAvailable(sc.GetService(), brand) {
 			continue
 		}
-		if domainSet[sc.Service] && shortcutSupportsIdentity(sc, identity) {
+		if domainSet[sc.GetService()] && shortcutSupportsIdentity(sc, identity) {
 			for _, s := range sc.DeclaredScopesForIdentity(identity) {
 				scopeSet[s] = true
 			}
@@ -561,11 +557,11 @@ func allKnownDomains(brand core.LarkBrand) map[string]bool {
 		}
 	}
 	for _, sc := range shortcuts.AllShortcuts() {
-		if !shortcuts.IsShortcutServiceAvailable(sc.Service, brand) {
+		if !shortcuts.IsShortcutServiceAvailable(sc.GetService(), brand) {
 			continue
 		}
-		if !registry.HasAuthDomain(sc.Service) {
-			domains[sc.Service] = true
+		if !registry.HasAuthDomain(sc.GetService()) {
+			domains[sc.GetService()] = true
 		}
 	}
 	return domains
@@ -584,8 +580,8 @@ func sortedKnownDomains(brand core.LarkBrand) []string {

 // shortcutSupportsIdentity checks if a shortcut supports the given identity ("user" or "bot").
 // Empty AuthTypes defaults to ["user"].
-func shortcutSupportsIdentity(sc common.Shortcut, identity string) bool {
-	authTypes := sc.AuthTypes
+func shortcutSupportsIdentity(sc common.ShortcutDescriptor, identity string) bool {
+	authTypes := sc.GetAuthTypes()
 	if len(authTypes) == 0 {
 		authTypes = []string{"user"}
 	}
--- a/cmd/auth/login_interactive.go
+++ b/cmd/auth/login_interactive.go
@@ -64,12 +64,13 @@ func getDomainMetadata(lang string) []domainMeta {
 		shortcutOnlySet[n] = true
 	}
 	for _, sc := range shortcuts.AllShortcuts() {
-		if !seen[sc.Service] {
-			if shortcutOnlySet[sc.Service] && !registry.HasAuthDomain(sc.Service) {
-				dm := buildDomainMeta(sc.Service, lang)
+		svc := sc.GetService()
+		if !seen[svc] {
+			if shortcutOnlySet[svc] && !registry.HasAuthDomain(svc) {
+				dm := buildDomainMeta(svc, lang)
 				domains = append(domains, dm)
 			}
-			seen[sc.Service] = true
+			seen[svc] = true
 		}
 	}

@@ -92,11 +93,16 @@ func buildDomainMeta(name, lang string) domainMeta {
 			Description: desc,
 		}
 	}
-	// Fallback: read from the typed service spec (legacy)
+	// Fallback: read from from_meta spec (legacy)
+	meta := registry.LoadFromMeta(name)
 	dm := domainMeta{Name: name}
-	if svc, ok := registry.ServiceTyped(name); ok {
-		dm.Title = svc.Title
-		dm.Description = svc.Description
+	if meta != nil {
+		if t, ok := meta["title"].(string); ok {
+			dm.Title = t
+		}
+		if d, ok := meta["description"].(string); ok {
+			dm.Description = d
+		}
 	}
 	return dm
 }
--- a/cmd/auth/login_messages.go
+++ b/cmd/auth/login_messages.go
@@ -128,5 +128,5 @@ func getLoginMsg(lang i18n.Lang) *loginMsg {
 // (not backed by from_meta service specs). Descriptions are now centralized in
 // service_descriptions.json.
 func getShortcutOnlyDomainNames() []string {
-	return []string{"base", "contact", "docs", "markdown", "apps", "note"}
+	return []string{"base", "contact", "docs", "markdown", "apps"}
 }
--- a/cmd/auth/login_test.go
+++ b/cmd/auth/login_test.go
@@ -9,7 +9,6 @@ import (
 	"errors"
 	"io"
 	"net/http"
-	"slices"
 	"sort"
 	"strings"
 	"testing"
@@ -99,7 +98,7 @@ func TestNormalizeScopeInput(t *testing.T) {

 func TestShortcutSupportsIdentity_DefaultUser(t *testing.T) {
 	// Empty AuthTypes defaults to ["user"]
-	sc := common.Shortcut{AuthTypes: nil}
+	sc := &common.Shortcut{AuthTypes: nil}
 	if !shortcutSupportsIdentity(sc, "user") {
 		t.Error("expected default to support 'user'")
 	}
@@ -109,7 +108,7 @@ func TestShortcutSupportsIdentity_DefaultUser(t *testing.T) {
 }

 func TestShortcutSupportsIdentity_ExplicitTypes(t *testing.T) {
-	sc := common.Shortcut{AuthTypes: []string{"user", "bot"}}
+	sc := &common.Shortcut{AuthTypes: []string{"user", "bot"}}
 	if !shortcutSupportsIdentity(sc, "user") {
 		t.Error("expected to support 'user'")
 	}
@@ -122,7 +121,7 @@ func TestShortcutSupportsIdentity_ExplicitTypes(t *testing.T) {
 }

 func TestShortcutSupportsIdentity_BotOnly(t *testing.T) {
-	sc := common.Shortcut{AuthTypes: []string{"bot"}}
+	sc := &common.Shortcut{AuthTypes: []string{"bot"}}
 	if shortcutSupportsIdentity(sc, "user") {
 		t.Error("expected bot-only to NOT support 'user'")
 	}
@@ -215,12 +214,6 @@ func TestGetShortcutOnlyDomainNames_HaveDescriptions(t *testing.T) {
 	}
 }

-func TestGetShortcutOnlyDomainNames_IncludesNote(t *testing.T) {
-	if !slices.Contains(getShortcutOnlyDomainNames(), "note") {
-		t.Fatal("shortcut-only domains must include note so auth login can select vc:note:read")
-	}
-}
-
 func TestCollectScopesForDomains(t *testing.T) {
 	projects := registry.ListFromMetaProjects()
 	if len(projects) == 0 {
@@ -878,7 +871,7 @@ func TestAuthLoginRun_DeviceCodeTokenNilCleansScopeCache(t *testing.T) {
 // contract that when --json is set and pollDeviceToken returns OK=false,
 // stdout carries the structured authorization_failed event and stderr is
 // NOT polluted with a typed envelope. The returned error is a bare
-// BareError with ExitAuth so the dispatcher only propagates the exit code
+// ExitError with ExitAuth so the dispatcher only propagates the exit code
 // without emitting a second envelope on top of the JSON event.
 func TestAuthLoginRun_JSONAbort_StdoutEventOnly_StderrEmpty(t *testing.T) {
 	keyring.MockInit()
@@ -945,13 +938,16 @@ func TestAuthLoginRun_JSONAbort_StdoutEventOnly_StderrEmpty(t *testing.T) {
 		t.Errorf("stderr should not contain JSON envelope fields, got: %s", stderrStr)
 	}

-	// Returned error must be the bare *output.BareError signal (no envelope).
-	var bareErr *output.BareError
-	if !errors.As(err, &bareErr) {
-		t.Fatalf("expected *output.BareError, got %T: %v", err, err)
+	// Returned error must be the bare *output.ExitError signal (no envelope).
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) {
+		t.Fatalf("expected *output.ExitError, got %T: %v", err, err)
 	}
-	if bareErr.Code != output.ExitAuth {
-		t.Fatalf("BareError.Code = %d, want %d", bareErr.Code, output.ExitAuth)
+	if exitErr.Code != output.ExitAuth {
+		t.Fatalf("ExitError.Code = %d, want %d", exitErr.Code, output.ExitAuth)
+	}
+	if exitErr.Detail != nil {
+		t.Errorf("ExitError.Detail should be nil for bare signal, got: %+v", exitErr.Detail)
 	}
 }

--- a/cmd/auth/logout.go
+++ b/cmd/auth/logout.go
@@ -18,7 +18,6 @@ import (
 // LogoutOptions holds all inputs for auth logout.
 type LogoutOptions struct {
 	Factory *cmdutil.Factory
-	JSON    bool
 }

 // NewCmdAuthLogout creates the auth logout subcommand.
@@ -35,7 +34,6 @@ func NewCmdAuthLogout(f *cmdutil.Factory, runF func(*LogoutOptions) error) *cobr
 			return authLogoutRun(opts)
 		},
 	}
-	cmd.Flags().BoolVar(&opts.JSON, "json", false, "structured JSON output")
 	cmdutil.SetRisk(cmd, "write")

 	return cmd
@@ -46,65 +44,25 @@ func authLogoutRun(opts *LogoutOptions) error {

 	multi, _ := core.LoadMultiAppConfig()
 	if multi == nil || len(multi.Apps) == 0 {
-		if opts.JSON {
-			output.PrintJson(f.IOStreams.Out, map[string]interface{}{
-				"ok":        true,
-				"loggedOut": false,
-				"reason":    "not_configured",
-			})
-			return nil
-		}
 		fmt.Fprintln(f.IOStreams.ErrOut, "No configuration found.")
 		return nil
 	}

 	app := multi.CurrentAppConfig(f.Invocation.Profile)
 	if app == nil || len(app.Users) == 0 {
-		if opts.JSON {
-			output.PrintJson(f.IOStreams.Out, map[string]interface{}{
-				"ok":        true,
-				"loggedOut": false,
-				"reason":    "not_logged_in",
-			})
-			return nil
-		}
 		fmt.Fprintln(f.IOStreams.ErrOut, "Not logged in.")
 		return nil
 	}

-	httpClient, httpErr := f.HttpClient()
-	appSecret, secretErr := core.ResolveSecretInput(app.AppSecret, f.Keychain)
-
 	for _, user := range app.Users {
-		if httpErr == nil && secretErr == nil {
-			if token := larkauth.GetStoredToken(app.AppId, user.UserOpenId); token != nil {
-				revokeToken := token.RefreshToken
-				tokenTypeHint := "refresh_token"
-				if revokeToken == "" {
-					revokeToken = token.AccessToken
-					tokenTypeHint = "access_token"
-				}
-				if revokeToken != "" {
-					_ = larkauth.RevokeToken(httpClient, app.AppId, appSecret, app.Brand, revokeToken, tokenTypeHint)
-				}
-			}
-		}
 		if err := larkauth.RemoveStoredToken(app.AppId, user.UserOpenId); err != nil {
 			fmt.Fprintf(f.IOStreams.ErrOut, "Warning: failed to remove token for %s: %v\n", user.UserOpenId, err)
 		}
 	}
-
 	app.Users = []core.AppUser{}
 	if err := core.SaveMultiAppConfig(multi); err != nil {
 		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
 	}
-	if opts.JSON {
-		output.PrintJson(f.IOStreams.Out, map[string]interface{}{
-			"ok":        true,
-			"loggedOut": true,
-		})
-		return nil
-	}
 	output.PrintSuccess(f.IOStreams.ErrOut, "Logged out")
 	return nil
 }
--- a/cmd/auth/logout_test.go
+++ b/cmd/auth/logout_test.go
@@ -1,356 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package auth
-
-import (
-	"encoding/json"
-	"net/url"
-	"strings"
-	"testing"
-
-	larkauth "github.com/larksuite/cli/internal/auth"
-	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/larksuite/cli/internal/core"
-	"github.com/larksuite/cli/internal/httpmock"
-	"github.com/zalando/go-keyring"
-)
-
-func writeLogoutConfig(t *testing.T, users []core.AppUser) {
-	t.Helper()
-	if err := core.SaveMultiAppConfig(&core.MultiAppConfig{
-		CurrentApp: "test-app",
-		Apps: []core.AppConfig{
-			{
-				AppId:     "test-app",
-				AppSecret: core.PlainSecret("test-secret"),
-				Brand:     core.BrandFeishu,
-				Users:     users,
-			},
-		},
-	}); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-}
-
-func TestAuthLogoutRun_JSONMode_NotConfigured_WritesStdoutOnly(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authLogoutRun(&LogoutOptions{Factory: f, JSON: true}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	var payload map[string]any
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("stdout must be valid JSON: %v\nstdout=%s", err, stdout.String())
-	}
-	if payload["ok"] != true {
-		t.Errorf("stdout.ok = %v, want true", payload["ok"])
-	}
-	if payload["loggedOut"] != false {
-		t.Errorf("stdout.loggedOut = %v, want false", payload["loggedOut"])
-	}
-	if payload["reason"] != "not_configured" {
-		t.Errorf("stdout.reason = %v, want not_configured", payload["reason"])
-	}
-	if stderr.Len() != 0 {
-		t.Errorf("stderr must stay empty in JSON mode, got:\n%s", stderr.String())
-	}
-}
-
-func TestAuthLogoutRun_JSONMode_NotLoggedIn_WritesStdoutOnly(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	writeLogoutConfig(t, nil)
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authLogoutRun(&LogoutOptions{Factory: f, JSON: true}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	var payload map[string]any
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("stdout must be valid JSON: %v\nstdout=%s", err, stdout.String())
-	}
-	if payload["ok"] != true {
-		t.Errorf("stdout.ok = %v, want true", payload["ok"])
-	}
-	if payload["loggedOut"] != false {
-		t.Errorf("stdout.loggedOut = %v, want false", payload["loggedOut"])
-	}
-	if payload["reason"] != "not_logged_in" {
-		t.Errorf("stdout.reason = %v, want not_logged_in", payload["reason"])
-	}
-	if stderr.Len() != 0 {
-		t.Errorf("stderr must stay empty in JSON mode, got:\n%s", stderr.String())
-	}
-}
-
-func TestAuthLogoutRun_JSONMode_Success_WritesStdoutOnly(t *testing.T) {
-	keyring.MockInit()
-	t.Setenv("HOME", t.TempDir())
-	t.Setenv("LARKSUITE_CLI_DATA_DIR", t.TempDir())
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	writeLogoutConfig(t, []core.AppUser{{UserOpenId: "ou_user", UserName: "tester"}})
-	if err := larkauth.SetStoredToken(&larkauth.StoredUAToken{
-		AppId:      "test-app",
-		UserOpenId: "ou_user",
-	}); err != nil {
-		t.Fatalf("SetStoredToken() error = %v", err)
-	}
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authLogoutRun(&LogoutOptions{Factory: f, JSON: true}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	var payload map[string]any
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("stdout must be valid JSON: %v\nstdout=%s", err, stdout.String())
-	}
-	if payload["ok"] != true {
-		t.Errorf("stdout.ok = %v, want true", payload["ok"])
-	}
-	if payload["loggedOut"] != true {
-		t.Errorf("stdout.loggedOut = %v, want true", payload["loggedOut"])
-	}
-	if _, hasReason := payload["reason"]; hasReason {
-		t.Errorf("stdout.reason must be absent on success, got %v", payload["reason"])
-	}
-	if stderr.Len() != 0 {
-		t.Errorf("stderr must stay empty in JSON mode, got:\n%s", stderr.String())
-	}
-}
-
-func TestAuthLogoutRun_DefaultMode_KeepsTextOutput(t *testing.T) {
-	keyring.MockInit()
-	t.Setenv("HOME", t.TempDir())
-	t.Setenv("LARKSUITE_CLI_DATA_DIR", t.TempDir())
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	writeLogoutConfig(t, []core.AppUser{{UserOpenId: "ou_user", UserName: "tester"}})
-	if err := larkauth.SetStoredToken(&larkauth.StoredUAToken{
-		AppId:      "test-app",
-		UserOpenId: "ou_user",
-	}); err != nil {
-		t.Fatalf("SetStoredToken() error = %v", err)
-	}
-
-	f, stdout, stderr, _ := cmdutil.TestFactory(t, nil)
-	if err := authLogoutRun(&LogoutOptions{Factory: f}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	if stdout.Len() != 0 {
-		t.Errorf("stdout must stay empty in default mode, got:\n%s", stdout.String())
-	}
-	if !strings.Contains(stderr.String(), "Logged out") {
-		t.Errorf("stderr = %q, want success text", stderr.String())
-	}
-}
-
-func TestAuthLogoutRun_RevokesTokenAndClearsLocalState(t *testing.T) {
-	keyring.MockInit()
-	setupLoginConfigDir(t)
-	t.Setenv("HOME", t.TempDir())
-
-	multi := &core.MultiAppConfig{
-		CurrentApp: "default",
-		Apps: []core.AppConfig{
-			{
-				Name:      "default",
-				AppId:     "cli_test",
-				AppSecret: core.PlainSecret("secret"),
-				Brand:     core.BrandFeishu,
-				Users:     []core.AppUser{{UserOpenId: "ou_user", UserName: "tester"}},
-			},
-		},
-	}
-	if err := core.SaveMultiAppConfig(multi); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-	if err := larkauth.SetStoredToken(&larkauth.StoredUAToken{
-		AppId:        "cli_test",
-		UserOpenId:   "ou_user",
-		AccessToken:  "user-access-token",
-		RefreshToken: "user-refresh-token",
-	}); err != nil {
-		t.Fatalf("SetStoredToken() error = %v", err)
-	}
-
-	f, _, stderr, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		ProfileName: "default",
-		AppID:       "cli_test",
-		AppSecret:   "secret",
-		Brand:       core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		Method: "POST",
-		URL:    larkauth.PathOAuthRevoke,
-		Body:   map[string]interface{}{"code": 0},
-		BodyFilter: func(body []byte) bool {
-			values, err := url.ParseQuery(string(body))
-			if err != nil {
-				return false
-			}
-			return values.Get("client_id") == "cli_test" &&
-				values.Get("client_secret") == "secret" &&
-				values.Get("token") == "user-refresh-token" &&
-				values.Get("token_type_hint") == "refresh_token"
-		},
-	})
-
-	if err := authLogoutRun(&LogoutOptions{Factory: f}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	if got := stderr.String(); !strings.Contains(got, "Logged out") {
-		t.Fatalf("stderr = %q, want Logged out", got)
-	}
-	if got := larkauth.GetStoredToken("cli_test", "ou_user"); got != nil {
-		t.Fatalf("expected stored token removed, got %#v", got)
-	}
-	saved, err := core.LoadMultiAppConfig()
-	if err != nil {
-		t.Fatalf("LoadMultiAppConfig() error = %v", err)
-	}
-	if len(saved.Apps) != 1 || len(saved.Apps[0].Users) != 0 {
-		t.Fatalf("expected users cleared, got %#v", saved.Apps)
-	}
-}
-
-func TestAuthLogoutRun_FallsBackToAccessTokenWhenRefreshTokenMissing(t *testing.T) {
-	keyring.MockInit()
-	setupLoginConfigDir(t)
-	t.Setenv("HOME", t.TempDir())
-
-	multi := &core.MultiAppConfig{
-		CurrentApp: "default",
-		Apps: []core.AppConfig{
-			{
-				Name:      "default",
-				AppId:     "cli_test",
-				AppSecret: core.PlainSecret("secret"),
-				Brand:     core.BrandFeishu,
-				Users:     []core.AppUser{{UserOpenId: "ou_user", UserName: "tester"}},
-			},
-		},
-	}
-	if err := core.SaveMultiAppConfig(multi); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-	if err := larkauth.SetStoredToken(&larkauth.StoredUAToken{
-		AppId:       "cli_test",
-		UserOpenId:  "ou_user",
-		AccessToken: "user-access-token",
-	}); err != nil {
-		t.Fatalf("SetStoredToken() error = %v", err)
-	}
-
-	f, _, stderr, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		ProfileName: "default",
-		AppID:       "cli_test",
-		AppSecret:   "secret",
-		Brand:       core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		Method: "POST",
-		URL:    larkauth.PathOAuthRevoke,
-		Body:   map[string]interface{}{"code": 0},
-		BodyFilter: func(body []byte) bool {
-			values, err := url.ParseQuery(string(body))
-			if err != nil {
-				return false
-			}
-			return values.Get("client_id") == "cli_test" &&
-				values.Get("client_secret") == "secret" &&
-				values.Get("token") == "user-access-token" &&
-				values.Get("token_type_hint") == "access_token"
-		},
-	})
-
-	if err := authLogoutRun(&LogoutOptions{Factory: f}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	if got := stderr.String(); !strings.Contains(got, "Logged out") {
-		t.Fatalf("stderr = %q, want Logged out", got)
-	}
-	if got := larkauth.GetStoredToken("cli_test", "ou_user"); got != nil {
-		t.Fatalf("expected stored token removed, got %#v", got)
-	}
-	saved, err := core.LoadMultiAppConfig()
-	if err != nil {
-		t.Fatalf("LoadMultiAppConfig() error = %v", err)
-	}
-	if len(saved.Apps) != 1 || len(saved.Apps[0].Users) != 0 {
-		t.Fatalf("expected users cleared, got %#v", saved.Apps)
-	}
-}
-
-func TestAuthLogoutRun_RevokeFailureStillClearsLocalState(t *testing.T) {
-	keyring.MockInit()
-	setupLoginConfigDir(t)
-	t.Setenv("HOME", t.TempDir())
-
-	multi := &core.MultiAppConfig{
-		CurrentApp: "default",
-		Apps: []core.AppConfig{
-			{
-				Name:      "default",
-				AppId:     "cli_test",
-				AppSecret: core.PlainSecret("secret"),
-				Brand:     core.BrandFeishu,
-				Users:     []core.AppUser{{UserOpenId: "ou_user", UserName: "tester"}},
-			},
-		},
-	}
-	if err := core.SaveMultiAppConfig(multi); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-	if err := larkauth.SetStoredToken(&larkauth.StoredUAToken{
-		AppId:        "cli_test",
-		UserOpenId:   "ou_user",
-		AccessToken:  "user-access-token",
-		RefreshToken: "user-refresh-token",
-	}); err != nil {
-		t.Fatalf("SetStoredToken() error = %v", err)
-	}
-
-	f, _, stderr, reg := cmdutil.TestFactory(t, &core.CliConfig{
-		ProfileName: "default",
-		AppID:       "cli_test",
-		AppSecret:   "secret",
-		Brand:       core.BrandFeishu,
-	})
-
-	reg.Register(&httpmock.Stub{
-		Method: "POST",
-		URL:    larkauth.PathOAuthRevoke,
-		Status: 500,
-		Body:   map[string]interface{}{"error": "server_error"},
-	})
-
-	if err := authLogoutRun(&LogoutOptions{Factory: f}); err != nil {
-		t.Fatalf("authLogoutRun() error = %v", err)
-	}
-
-	gotErr := stderr.String()
-	if strings.Contains(gotErr, "failed to revoke token for ou_user") {
-		t.Fatalf("stderr = %q, want no revoke warning", gotErr)
-	}
-	if !strings.Contains(gotErr, "Logged out") {
-		t.Fatalf("stderr = %q, want Logged out", gotErr)
-	}
-	if got := larkauth.GetStoredToken("cli_test", "ou_user"); got != nil {
-		t.Fatalf("expected stored token removed, got %#v", got)
-	}
-	saved, err := core.LoadMultiAppConfig()
-	if err != nil {
-		t.Fatalf("LoadMultiAppConfig() error = %v", err)
-	}
-	if len(saved.Apps) != 1 || len(saved.Apps[0].Users) != 0 {
-		t.Fatalf("expected users cleared, got %#v", saved.Apps)
-	}
-}
--- a/cmd/auth/scopes.go
+++ b/cmd/auth/scopes.go
@@ -19,7 +19,6 @@ type ScopesOptions struct {
 	Factory *cmdutil.Factory
 	Ctx     context.Context
 	Format  string
-	JSON    bool
 }

 // NewCmdAuthScopes creates the auth scopes subcommand.
@@ -31,9 +30,6 @@ func NewCmdAuthScopes(f *cmdutil.Factory, runF func(*ScopesOptions) error) *cobr
 		Short: "Query scopes enabled for the app",
 		RunE: func(cmd *cobra.Command, args []string) error {
 			opts.Ctx = cmd.Context()
-			if opts.JSON {
-				opts.Format = "json"
-			}
 			if runF != nil {
 				return runF(opts)
 			}
@@ -42,7 +38,6 @@ func NewCmdAuthScopes(f *cmdutil.Factory, runF func(*ScopesOptions) error) *cobr
 	}

 	cmd.Flags().StringVar(&opts.Format, "format", "json", "output format: json (default) | pretty")
-	cmd.Flags().BoolVar(&opts.JSON, "json", false, "structured JSON output")
 	cmdutil.SetRisk(cmd, "read")

 	return cmd
--- a/cmd/auth/status.go
+++ b/cmd/auth/status.go
@@ -17,7 +17,6 @@ import (
 type StatusOptions struct {
 	Factory *cmdutil.Factory
 	Verify  bool
-	JSON    bool
 }

 // NewCmdAuthStatus creates the auth status subcommand.
@@ -36,7 +35,6 @@ func NewCmdAuthStatus(f *cmdutil.Factory, runF func(*StatusOptions) error) *cobr
 	}

 	cmd.Flags().BoolVar(&opts.Verify, "verify", false, "verify token against server (requires network)")
-	cmd.Flags().BoolVar(&opts.JSON, "json", false, "structured JSON output")
 	cmdutil.SetRisk(cmd, "read")

 	return cmd
--- a/cmd/build.go
+++ b/cmd/build.go
@@ -6,7 +6,6 @@ package cmd
 import (
 	"context"
 	"io"
-	"io/fs"

 	"github.com/larksuite/cli/cmd/api"
 	"github.com/larksuite/cli/cmd/auth"
@@ -17,11 +16,8 @@ import (
 	"github.com/larksuite/cli/cmd/profile"
 	"github.com/larksuite/cli/cmd/schema"
 	"github.com/larksuite/cli/cmd/service"
-	"github.com/larksuite/cli/cmd/skill"
 	cmdupdate "github.com/larksuite/cli/cmd/update"
-	"github.com/larksuite/cli/cmd/whoami"
 	_ "github.com/larksuite/cli/events"
-	"github.com/larksuite/cli/internal/apicatalog"
 	"github.com/larksuite/cli/internal/build"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
@@ -35,13 +31,9 @@ import (
 type BuildOption func(*buildConfig)

 type buildConfig struct {
-	streams        *cmdutil.IOStreams
-	keychain       keychain.KeychainAccess
-	globals        GlobalOptions
-	skipPlugins    bool
-	skipStrictMode bool
-	skipService    bool
-	serviceCatalog *apicatalog.Catalog
+	streams  *cmdutil.IOStreams
+	keychain keychain.KeychainAccess
+	globals  GlobalOptions
 }

 // WithIO sets the IO streams for the CLI by wrapping raw reader/writers.
@@ -59,18 +51,6 @@ func WithKeychain(kc keychain.KeychainAccess) BuildOption {
 	}
 }

-// embeddedSkillContent is the skill tree wired into cmdutil.Factory.SkillContent
-// at build time. It is registered by the repo-root package main's init via
-// SetEmbeddedSkillContent — it cannot be threaded through main.go without
-// breaking the single-file preview build (see skills_embed.go). nil in builds
-// that embed no skills; the `skills` commands then return a typed internal error.
-var embeddedSkillContent fs.FS
-
-// SetEmbeddedSkillContent registers the embedded skill tree. Called from the
-// repo-root package main's init; a wrapper main can call it before Execute to
-// supply its own skill content.
-func SetEmbeddedSkillContent(fsys fs.FS) { embeddedSkillContent = fsys }
-
 // HideProfile sets the visibility policy for the root-level --profile flag.
 // When hide is true the flag stays registered (so existing invocations still
 // parse) but is omitted from help and shell completion. Typically called as
@@ -81,41 +61,6 @@ func HideProfile(hide bool) BuildOption {
 	}
 }

-// WithoutPlugins builds only repository-owned commands. It is intended for
-// inspection tools that need a deterministic command tree.
-func WithoutPlugins() BuildOption {
-	return func(c *buildConfig) {
-		c.skipPlugins = true
-	}
-}
-
-// WithoutStrictMode builds the complete repository-owned command tree without
-// applying user/profile strict-mode pruning. It is intended for offline
-// inspection tools, not production execution.
-func WithoutStrictMode() BuildOption {
-	return func(c *buildConfig) {
-		c.skipStrictMode = true
-	}
-}
-
-// WithoutServiceCommands builds only hand-authored commands. It is intended for
-// repository quality gates that should not depend on the remote OpenAPI
-// metadata command surface.
-func WithoutServiceCommands() BuildOption {
-	return func(c *buildConfig) {
-		c.skipService = true
-	}
-}
-
-// WithServiceCatalog builds generated service commands from a specific metadata
-// catalog. It is intended for offline inspection tools that need deterministic
-// embedded metadata while production execution keeps using the runtime catalog.
-func WithServiceCatalog(catalog apicatalog.Catalog) BuildOption {
-	return func(c *buildConfig) {
-		c.serviceCatalog = &catalog
-	}
-}
-
 // Build constructs the full command tree. It also installs registered
 // plugins and emits the Startup lifecycle event during assembly --
 // so Plugin.On(Startup) handlers run even if the returned command is
@@ -158,7 +103,6 @@ func buildInternal(ctx context.Context, inv cmdutil.InvocationContext, opts ...B
 	if cfg.keychain != nil {
 		f.Keychain = cfg.keychain
 	}
-	f.SkillContent = embeddedSkillContent
 	rootCmd := &cobra.Command{
 		Use:     "lark-cli",
 		Short:   "Lark/Feishu CLI — OAuth authorization, UAT management, API calls",
@@ -171,19 +115,8 @@ func buildInternal(ctx context.Context, inv cmdutil.InvocationContext, opts ...B
 	rootCmd.SetOut(cfg.streams.Out)
 	rootCmd.SetErr(cfg.streams.ErrOut)

-	// Root-only usage template (curated Usage synopsis + skills footer); see
-	// rootUsageTemplate.
-	rootCmd.SetUsageTemplate(rootUsageTemplate)
-
 	installTipsHelpFunc(rootCmd)
 	rootCmd.SilenceErrors = true
-	// SilenceUsage as a static field (not only in PersistentPreRun) so it also
-	// covers flag-parse errors, which fail before PreRun runs — otherwise cobra
-	// dumps usage instead of our structured error. SetFlagErrorFunc on root is
-	// inherited by every subcommand, turning unknown-flag errors into a
-	// structured "did you mean" envelope.
-	rootCmd.SilenceUsage = true
-	rootCmd.SetFlagErrorFunc(flagDidYouMean)

 	RegisterGlobalFlags(rootCmd.PersistentFlags(), &cfg.globals)
 	rootCmd.PersistentPreRun = func(cmd *cobra.Command, args []string) {
@@ -195,38 +128,20 @@ func buildInternal(ctx context.Context, inv cmdutil.InvocationContext, opts ...B
 	rootCmd.AddCommand(auth.NewCmdAuth(f))
 	rootCmd.AddCommand(profile.NewCmdProfile(f))
 	rootCmd.AddCommand(doctor.NewCmdDoctor(f))
-	rootCmd.AddCommand(whoami.NewCmdWhoami(f))
 	rootCmd.AddCommand(api.NewCmdApiWithContext(ctx, f, nil))
 	rootCmd.AddCommand(schema.NewCmdSchema(f, nil))
 	rootCmd.AddCommand(completion.NewCmdCompletion(f))
 	rootCmd.AddCommand(cmdupdate.NewCmdUpdate(f))
 	rootCmd.AddCommand(cmdevent.NewCmdEvents(f))
-	rootCmd.AddCommand(skill.NewCmdSkill(f))
-	if !cfg.skipService {
-		if cfg.serviceCatalog != nil {
-			service.RegisterServiceCommandsFromCatalog(ctx, rootCmd, f, *cfg.serviceCatalog)
-		} else {
-			service.RegisterServiceCommandsWithContext(ctx, rootCmd, f)
-		}
-	}
+	service.RegisterServiceCommandsWithContext(ctx, rootCmd, f)
 	shortcuts.RegisterShortcutsWithContext(ctx, rootCmd, f)

-	groupRootCommands(rootCmd)
-
 	installUnknownSubcommandGuard(rootCmd)
-	// Bare `lark-cli` in an interactive terminal offers an interactive upgrade
-	// before printing help; non-bare invocations and non-TTY are unaffected.
-	installRootUpgradePrompt(f, rootCmd)

-	if mode := f.ResolveStrictMode(ctx); mode.IsActive() && !cfg.skipStrictMode {
+	if mode := f.ResolveStrictMode(ctx); mode.IsActive() {
 		pruneForStrictMode(rootCmd, mode)
 	}

-	if cfg.skipPlugins {
-		recordInventory(nil)
-		return f, rootCmd, nil
-	}
-
 	installResult, installErr := installPluginsAndHooks(cfg.streams.ErrOut)
 	if installErr != nil {
 		installPluginInstallErrorGuard(rootCmd, installErr)
--- a/cmd/build_test.go
+++ b/cmd/build_test.go
@@ -1,46 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd
-
-import (
-	"context"
-	"strings"
-	"testing"
-
-	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/spf13/cobra"
-)
-
-func TestBuildWithoutPluginsStillBuildsBuiltinCommands(t *testing.T) {
-	root := Build(context.Background(), cmdutil.InvocationContext{}, WithoutPlugins())
-
-	if root == nil {
-		t.Fatal("Build returned nil root")
-	}
-	if findCommand(root, "api") == nil {
-		t.Fatal("builtin api command missing")
-	}
-	if findCommand(root, "docs +fetch") == nil {
-		t.Fatal("builtin docs +fetch shortcut missing")
-	}
-}
-
-func findCommand(root *cobra.Command, path string) *cobra.Command {
-	parts := strings.Fields(path)
-	cmd := root
-	for _, part := range parts {
-		var next *cobra.Command
-		for _, child := range cmd.Commands() {
-			if child.Name() == part {
-				next = child
-				break
-			}
-		}
-		if next == nil {
-			return nil
-		}
-		cmd = next
-	}
-	return cmd
-}
--- a/cmd/cmdexample_catalog_test.go
+++ b/cmd/cmdexample_catalog_test.go
@@ -1,160 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd_test
-
-import (
-	"sort"
-	"strings"
-)
-
-// universalFlags are accepted by every command (cobra auto-injects help; the
-// root injects version). They are never reported as unknown.
-var universalFlags = map[string]bool{"--help": true, "-h": true, "--version": true}
-
-// catalog is the source-of-truth command catalog: command path -> accepted flag
-// tokens. A path is the command words WITHOUT the "lark-cli" root prefix, e.g.
-// "contact +search-user". The root command is the empty path "".
-type catalog struct {
-	flagsByPath map[string]map[string]bool
-	group       map[string]bool // paths that are parent groups (have subcommands)
-	sorted      []string        // cached sorted paths for suggestCommand; invalidated on addCommand
-}
-
-func newCatalog() *catalog {
-	return &catalog{
-		flagsByPath: map[string]map[string]bool{},
-		group:       map[string]bool{},
-	}
-}
-
-// setGroup records whether path is a parent group (has subcommands). Leftover
-// words after a group node are unknown subcommands; after a leaf they are
-// positionals (e.g. "api GET /path").
-func (c *catalog) setGroup(path string, isGroup bool) {
-	if isGroup {
-		c.group[path] = true
-	}
-}
-
-func (c *catalog) isGroup(path string) bool { return c.group[path] }
-
-// addCommand registers a command path and the flags it accepts. Repeated calls
-// for the same path union the flag sets. flags are full tokens ("--query", "-q").
-func (c *catalog) addCommand(path string, flags []string) {
-	set := c.flagsByPath[path]
-	if set == nil {
-		set = map[string]bool{}
-		c.flagsByPath[path] = set
-	}
-	for _, f := range flags {
-		set[f] = true
-	}
-	c.sorted = nil // invalidate cached suggestion list
-}
-
-func (c *catalog) hasCommand(path string) bool {
-	_, ok := c.flagsByPath[path]
-	return ok
-}
-
-// hasFlag reports whether flag is accepted by command path (universal flags
-// always pass).
-func (c *catalog) hasFlag(path, flag string) bool {
-	if universalFlags[flag] {
-		return true
-	}
-	set := c.flagsByPath[path]
-	return set[flag]
-}
-
-// longestPrefix returns the longest known command path that is a prefix of
-// words, plus how many words it consumed. This separates real subcommands from
-// trailing positionals (e.g. "api GET /path" resolves to "api"). When words is
-// empty it falls back to the root command. ok=false means not even the first
-// word names a command.
-func (c *catalog) longestPrefix(words []string) (path string, n int, ok bool) {
-	if len(words) == 0 {
-		if c.hasCommand("") {
-			return "", 0, true
-		}
-		return "", 0, false
-	}
-	for i := len(words); i >= 1; i-- {
-		cand := strings.Join(words[:i], " ")
-		if c.hasCommand(cand) {
-			return cand, i, true
-		}
-	}
-	return "", 0, false
-}
-
-// paths returns all known command paths, sorted.
-func (c *catalog) paths() []string {
-	out := make([]string, 0, len(c.flagsByPath))
-	for p := range c.flagsByPath {
-		out = append(out, p)
-	}
-	sort.Strings(out)
-	return out
-}
-
-// suggestCommand returns the known command path closest to want (small edit
-// distance), for error hints. Returns "" when nothing is reasonably close.
-func (c *catalog) suggestCommand(want string) string {
-	if c.sorted == nil {
-		c.sorted = c.paths() // built once after the catalog is fully populated
-	}
-	return closest(want, c.sorted)
-}
-
-// suggestFlag returns the flag of path closest to flag, for error hints.
-func (c *catalog) suggestFlag(path, flag string) string {
-	set := c.flagsByPath[path]
-	cands := make([]string, 0, len(set))
-	for f := range set {
-		cands = append(cands, f)
-	}
-	sort.Strings(cands)
-	return closest(flag, cands)
-}
-
-// closest returns the candidate with the smallest Levenshtein distance to want,
-// but only if that distance is within a tolerance scaled to want's length
-// (avoids absurd suggestions).
-func closest(want string, cands []string) string {
-	best := ""
-	bestD := 1 << 30
-	for _, cand := range cands {
-		d := levenshtein(want, cand)
-		if d < bestD {
-			bestD, best = d, cand
-		}
-	}
-	tol := len(want)/2 + 1
-	if bestD > tol {
-		return ""
-	}
-	return best
-}
-
-func levenshtein(a, b string) int {
-	ra, rb := []rune(a), []rune(b)
-	prev := make([]int, len(rb)+1)
-	for j := range prev {
-		prev[j] = j
-	}
-	for i := 1; i <= len(ra); i++ {
-		cur := make([]int, len(rb)+1)
-		cur[0] = i
-		for j := 1; j <= len(rb); j++ {
-			cost := 1
-			if ra[i-1] == rb[j-1] {
-				cost = 0
-			}
-			cur[j] = min(prev[j]+1, cur[j-1]+1, prev[j-1]+cost)
-		}
-		prev = cur
-	}
-	return prev[len(rb)]
-}
--- a/cmd/cmdexample_check_test.go
+++ b/cmd/cmdexample_check_test.go
@@ -1,60 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd_test
-
-import "strings"
-
-// Finding kinds.
-const (
-	unknownCommand = "unknown_command"
-	unknownFlag    = "unknown_flag"
-)
-
-// finding is a single mismatch between an example command reference and the
-// catalog.
-type finding struct {
-	line    int
-	raw     string
-	kind    string // unknownCommand | unknownFlag
-	path    string // resolved command path (unknownFlag) or attempted path (unknownCommand)
-	flag    string // offending flag (unknownFlag only)
-	suggest string // nearest known command/flag, "" if none close
-}
-
-// checkRefs validates refs against cat and returns all mismatches in order.
-func checkRefs(cat *catalog, refs []ref) []finding {
-	var out []finding
-	for _, r := range refs {
-		path, n, ok := cat.longestPrefix(r.words)
-		if !ok {
-			attempted := strings.Join(r.words, " ")
-			out = append(out, finding{
-				line: r.line, raw: r.raw, kind: unknownCommand,
-				path: attempted, suggest: cat.suggestCommand(attempted),
-			})
-			continue
-		}
-		// Leftover words after a group node are an unknown subcommand (e.g. a
-		// mistyped method like "batch_modify_message"). After a leaf they are
-		// positionals (e.g. "api GET /path"), so only groups trigger this.
-		if n < len(r.words) && cat.isGroup(path) {
-			attempted := strings.Join(r.words, " ")
-			out = append(out, finding{
-				line: r.line, raw: r.raw, kind: unknownCommand,
-				path: attempted, suggest: cat.suggestCommand(attempted),
-			})
-			continue
-		}
-		for _, f := range r.flags {
-			if cat.hasFlag(path, f) {
-				continue
-			}
-			out = append(out, finding{
-				line: r.line, raw: r.raw, kind: unknownFlag,
-				path: path, flag: f, suggest: cat.suggestFlag(path, f),
-			})
-		}
-	}
-	return out
-}
--- a/cmd/cmdexample_parse_test.go
+++ b/cmd/cmdexample_parse_test.go
@@ -1,222 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd_test
-
-import (
-	"regexp"
-	"strings"
-)
-
-// ref is one lark-cli command reference extracted from a shortcut example.
-type ref struct {
-	line  int      // 1-based line number (the line where the command starts)
-	raw   string   // reconstructed command text, for error display
-	words []string // command words before the first flag (subcommand candidates)
-	flags []string // flag tokens used, e.g. "--query", "-q"
-}
-
-const cliToken = "lark-cli"
-
-// subcommandStart guards against false positives from prose: a real command's
-// first word is ASCII (a service name or a +shortcut). A token starting with
-// CJK / punctuation is treated as narration, not a command.
-var subcommandStart = regexp.MustCompile(`^[A-Za-z+]`)
-
-// shellStops are standalone tokens that terminate a command (pipes, redirects,
-// separators). Separators glued to a token (`get;`, `foo|`) are handled inline.
-var shellStops = map[string]bool{
-	"|": true, "||": true, "&&": true, "&": true, ";": true,
-	">": true, ">>": true, "<": true, "2>": true, "2>&1": true,
-}
-
-// wordTrailPunct is sentence / CJK punctuation that can cling to a command word
-// in prose ("auth login." / "auth login，"); stripped so the word still resolves
-// instead of being dropped as an unknown command or non-ASCII narration.
-const wordTrailPunct = `.,;:!?"')]}，。、；：！？）】」』`
-
-// parseRefs extracts every lark-cli command reference from text (a shortcut's
-// Tips line, which may embed an "Example: lark-cli ..." command). It is
-// deliberately format-agnostic: it keys on the "lark-cli" token whether it sits
-// in a ```bash fence, an inline `code` span, or bare prose. Backslash
-// line-continuations are joined first so a multi-line invocation is parsed as
-// one command; inline-code backticks and trailing # comments terminate it.
-func parseRefs(content string) []ref {
-	var refs []ref
-	lines := strings.Split(content, "\n")
-	for i := 0; i < len(lines); i++ {
-		lineNo := i + 1
-		logical := lines[i]
-		// Shell line continuation: a trailing backslash joins the next physical
-		// line. Without this, flags on the continuation lines of a multi-line
-		// `lark-cli ... \` example are never seen by the checker.
-		for endsWithBackslash(logical) && i+1 < len(lines) {
-			logical = strings.TrimRight(logical, " \t")
-			logical = logical[:len(logical)-1] // drop the trailing backslash
-			i++
-			logical += " " + lines[i]
-		}
-		refs = append(refs, parseLine(logical, lineNo)...)
-	}
-	return refs
-}
-
-func endsWithBackslash(s string) bool {
-	return strings.HasSuffix(strings.TrimRight(s, " \t"), `\`)
-}
-
-func parseLine(line string, lineNo int) []ref {
-	var refs []ref
-	rest := line
-	for {
-		idx := strings.Index(rest, cliToken)
-		if idx < 0 {
-			break
-		}
-		after := rest[idx+len(cliToken):]
-		beforeOK := idx == 0 || isBoundary(rest[idx-1])
-		afterOK := after == "" || isBoundary(after[0])
-		if beforeOK && afterOK {
-			if words, flags, raw, ok := parseCmd(after); ok {
-				refs = append(refs, ref{line: lineNo, raw: cliToken + raw, words: words, flags: flags})
-			}
-		}
-		rest = after
-	}
-	return refs
-}
-
-// parseCmd tokenizes the text following "lark-cli" into leading command words
-// (the subcommand path, up to the first flag) and flag tokens. It stops at a
-// shell separator (standalone or glued), an inline-code backtick, a comment, or
-// a placeholder/prose word. ok=false filters out non-commands.
-func parseCmd(after string) (words, flags []string, raw string, ok bool) {
-	// An inline code span ends at the next backtick; a command never spans one.
-	if i := strings.IndexByte(after, '`'); i >= 0 {
-		after = after[:i]
-	}
-	// Drop $(...) command substitutions so flags belonging to the inner command
-	// (e.g. `--data "$(jq -n --arg x ...)"`) are not mistaken for lark-cli flags.
-	after = stripCmdSubst(after)
-
-	var kept []string
-	inFlags := false
-	for _, orig := range strings.Fields(after) {
-		tok := orig
-		if shellStops[tok] || strings.HasPrefix(tok, "#") {
-			break
-		}
-		// A shell separator glued to a token ends the command mid-token
-		// ("get;", "foo|next"): keep the part before it, handle it, then stop.
-		stop := false
-		if i := strings.IndexAny(tok, ";|"); i >= 0 {
-			tok, stop = tok[:i], true
-		}
-		switch {
-		case tok == "" || tok == "-":
-			// empty (after a glued separator) or a bare stdin marker — skip
-		case strings.HasPrefix(tok, "-"):
-			if f := normalizeFlag(tok); f != "" {
-				inFlags = true
-				flags = append(flags, f)
-				kept = append(kept, tok)
-			}
-		case inFlags:
-			// positional / flag value after the first flag — not a command word
-			kept = append(kept, tok)
-		default:
-			// Command-path word. ASCII placeholder markers (<x>, [x], {x|y},
-			// +<verb>, ...) end the command — checked on the RAW token so the
-			// trailing-punct stripping below cannot erase a "..." ellipsis
-			// ("base +..." must stay a placeholder, not become "+").
-			if strings.ContainsAny(tok, "<>[]{}|") || strings.Contains(tok, "...") {
-				stop = true
-				break
-			}
-			// Strip trailing sentence/CJK punctuation so "login." / "login，"
-			// resolve to "login"; non-ASCII narration ends the command.
-			w := strings.TrimRight(tok, wordTrailPunct)
-			if w == "" || hasNonASCII(w) {
-				stop = true
-				break
-			}
-			words = append(words, w)
-			kept = append(kept, tok)
-		}
-		if stop {
-			break
-		}
-	}
-	if len(kept) > 0 {
-		raw = " " + strings.Join(kept, " ")
-	}
-	// Keep root-only refs ("lark-cli --help") and refs whose first word looks
-	// like a subcommand; drop prose ("lark-cli 就能搞定 ...").
-	if len(words) == 0 {
-		return words, flags, raw, len(flags) > 0
-	}
-	if !subcommandStart.MatchString(words[0]) {
-		return nil, nil, "", false
-	}
-	return words, flags, raw, true
-}
-
-// stripCmdSubst removes $(...) command substitutions (including nested ones)
-// from s, leaving the surrounding text intact. Backtick substitutions are
-// already handled upstream (a command never spans a backtick).
-func stripCmdSubst(s string) string {
-	var b strings.Builder
-	depth := 0
-	for i := 0; i < len(s); i++ {
-		if depth == 0 && i+1 < len(s) && s[i] == '$' && s[i+1] == '(' {
-			depth = 1
-			i++ // skip '('
-			continue
-		}
-		if depth > 0 {
-			switch s[i] {
-			case '(':
-				depth++
-			case ')':
-				depth--
-			}
-			continue
-		}
-		b.WriteByte(s[i])
-	}
-	return b.String()
-}
-
-// isPlaceholderOrProse reports whether a command word is a doc placeholder
-// (<resource>, [flags], {a|b}, +<verb>, ...) or narration (CJK / other
-// non-ASCII), rather than a literal command token.
-func isPlaceholderOrProse(w string) bool {
-	if hasNonASCII(w) {
-		return true
-	}
-	return strings.ContainsAny(w, "<>[]{}|") || strings.Contains(w, "...")
-}
-
-func hasNonASCII(s string) bool {
-	return strings.IndexFunc(s, func(r rune) bool { return r > 127 }) >= 0
-}
-
-// flagShape matches the leading flag token, stripping any trailing junk such as
-// a "=value" suffix or punctuation that bled in from the surrounding markdown
-// ("--help\"", "--help;", "--params={}"). The underscore is allowed because
-// real flags use it ("--input_format", "--output_as"). Returns "" for non-flags.
-var flagShape = regexp.MustCompile(`^--?[A-Za-z][A-Za-z0-9_-]*`)
-
-// normalizeFlag extracts the canonical flag token from tok, or "" if tok is not
-// a real flag (e.g. a shell-string fragment like "-草稿'").
-func normalizeFlag(tok string) string {
-	return flagShape.FindString(tok)
-}
-
-func isBoundary(b byte) bool {
-	switch b {
-	case ' ', '\t', '`', '(', ')', '\'', '"', '*':
-		return true
-	}
-	return false
-}
--- a/cmd/cmdexample_test.go
+++ b/cmd/cmdexample_test.go
@@ -1,113 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-// This file and its cmdexample_*_test.go siblings implement a test-only check:
-// the example commands embedded in shortcut definitions (the "Example: lark-cli
-// ..." lines in each shortcut's Tips, shown in --help) must match the real
-// command tree. It lives entirely in _test.go files (package cmd_test) so it
-// ships in no binary and is not importable by product code; the truth source is
-// cmd.Build, the same tree the binary uses, so the check cannot drift.
-//
-// It runs in the standard unit-test CI job (go test ./cmd/...). A mismatch — an
-// example using a renamed command or an unaccepted flag — fails that job.
-
-package cmd_test
-
-import (
-	"context"
-	"sort"
-	"strings"
-	"testing"
-
-	"github.com/larksuite/cli/cmd"
-	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/larksuite/cli/shortcuts"
-	"github.com/spf13/cobra"
-	"github.com/spf13/pflag"
-)
-
-// TestShortcutExampleCommands checks the example commands embedded in every
-// shortcut's Tips against the live command tree. A shortcut that defines no
-// example is simply skipped.
-//
-// Because the examples and the command definitions live in the same Go code,
-// this is a self-consistency check: any mismatch (an example using a renamed
-// command or a flag the command doesn't accept) is a bug to fix at the source.
-// It runs over all shortcuts — no baseline, no diff — since a wrong example is
-// always a defect, never acceptable "pre-existing drift".
-func TestShortcutExampleCommands(t *testing.T) {
-	// Reproducibility: use the embedded API metadata (not a developer's stale
-	// ~/.lark-cli remote cache, which can miss commands) and an empty config
-	// dir so local strict mode / plugins / policy cannot reshape the tree.
-	// t.Setenv auto-restores after the test, so other cmd tests are unaffected.
-	t.Setenv("LARKSUITE_CLI_REMOTE_META", "off")
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-
-	cat := buildCmdExampleCatalog()
-
-	type located struct {
-		shortcut string
-		f        finding
-	}
-	var findings []located
-	for _, sc := range shortcuts.AllShortcuts() {
-		var refs []ref
-		for _, tip := range sc.Tips {
-			refs = append(refs, parseRefs(tip)...)
-		}
-		label := strings.TrimSpace(sc.Service + " " + sc.Command)
-		for _, f := range checkRefs(cat, refs) {
-			findings = append(findings, located{shortcut: label, f: f})
-		}
-	}
-
-	if len(findings) == 0 {
-		return
-	}
-	sort.Slice(findings, func(i, j int) bool { return findings[i].shortcut < findings[j].shortcut })
-	for _, lf := range findings {
-		hint := ""
-		if lf.f.suggest != "" {
-			hint = "  (did you mean " + lf.f.suggest + "?)"
-		}
-		if lf.f.kind == unknownFlag {
-			t.Errorf("shortcut %q example uses unknown flag %s on %q%s\n      %s",
-				lf.shortcut, lf.f.flag, lf.f.path, hint, strings.TrimSpace(lf.f.raw))
-		} else {
-			t.Errorf("shortcut %q example uses unknown command %q%s\n      %s",
-				lf.shortcut, lf.f.path, hint, strings.TrimSpace(lf.f.raw))
-		}
-	}
-	t.Fatalf("%d shortcut example command(s) don't match the real CLI — "+
-		"fix the Example in the shortcut definition.", len(findings))
-}
-
-// buildCmdExampleCatalog walks the live cobra command tree and records every
-// command path (minus the "lark-cli" root prefix) with its accepted flags and
-// whether it is a parent group. This is the same Build() the binary uses, so
-// the catalog can never drift from the real commands.
-func buildCmdExampleCatalog() *catalog {
-	root := cmd.Build(context.Background(), cmdutil.InvocationContext{})
-	cat := newCatalog()
-	var walk func(c *cobra.Command)
-	walk = func(c *cobra.Command) {
-		path := strings.TrimSpace(strings.TrimPrefix(c.CommandPath(), "lark-cli"))
-		var flags []string
-		add := func(fl *pflag.Flag) {
-			flags = append(flags, "--"+fl.Name)
-			if fl.Shorthand != "" {
-				flags = append(flags, "-"+fl.Shorthand)
-			}
-		}
-		c.Flags().VisitAll(add)
-		c.InheritedFlags().VisitAll(add)
-		c.PersistentFlags().VisitAll(add) // root's own persistent flags (e.g. --profile)
-		cat.addCommand(path, flags)
-		cat.setGroup(path, c.HasSubCommands())
-		for _, sub := range c.Commands() {
-			walk(sub)
-		}
-	}
-	walk(root)
-	return cat
-}
--- a/cmd/cmdexample_units_test.go
+++ b/cmd/cmdexample_units_test.go
@@ -1,233 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd_test
-
-import (
-	"strings"
-	"testing"
-)
-
-func testCatalog() *catalog {
-	c := newCatalog()
-	c.addCommand("", []string{"--profile"}) // root
-	c.setGroup("", true)
-	c.addCommand("contact", []string{"--profile"})
-	c.setGroup("contact", true)
-	c.addCommand("contact +search-user", []string{"--query", "--as", "--format", "-q"})
-	c.addCommand("api", []string{"--params", "--data", "--as"}) // leaf (no subcommands)
-	c.addCommand("mail", nil)
-	c.setGroup("mail", true)
-	c.addCommand("mail user_mailbox.messages", []string{"--profile"})
-	c.setGroup("mail user_mailbox.messages", true)
-	c.addCommand("mail user_mailbox.messages batch_modify", []string{"--params", "--data"})
-	return c
-}
-
-func TestCmdExampleCatalogHasCommandAndFlag(t *testing.T) {
-	c := testCatalog()
-	if !c.hasCommand("contact +search-user") {
-		t.Fatal("expected contact +search-user to exist")
-	}
-	if c.hasCommand("contact +nope") {
-		t.Fatal("did not expect contact +nope")
-	}
-	if !c.hasFlag("contact +search-user", "--query") {
-		t.Fatal("--query should be valid")
-	}
-	if c.hasFlag("contact +search-user", "--nope") {
-		t.Fatal("--nope should be invalid")
-	}
-	// universal flags pass on any command
-	for _, f := range []string{"--help", "-h", "--version"} {
-		if !c.hasFlag("contact +search-user", f) {
-			t.Fatalf("universal flag %s should pass", f)
-		}
-	}
-}
-
-func TestCmdExampleLongestPrefix(t *testing.T) {
-	c := testCatalog()
-	tests := []struct {
-		words  []string
-		want   string
-		wantN  int
-		wantOK bool
-	}{
-		{[]string{"contact", "+search-user"}, "contact +search-user", 2, true},
-		{[]string{"api", "GET", "/open-apis/x"}, "api", 1, true}, // trailing positionals
-		{[]string{"nope"}, "", 0, false},
-		{nil, "", 0, true}, // empty -> root
-	}
-	for _, tt := range tests {
-		got, n, ok := c.longestPrefix(tt.words)
-		if got != tt.want || n != tt.wantN || ok != tt.wantOK {
-			t.Errorf("longestPrefix(%v) = (%q,%d,%v), want (%q,%d,%v)",
-				tt.words, got, n, ok, tt.want, tt.wantN, tt.wantOK)
-		}
-	}
-}
-
-func refWordsOf(refs []ref) [][]string {
-	var out [][]string
-	for _, r := range refs {
-		out = append(out, r.words)
-	}
-	return out
-}
-
-func TestCmdExampleParseRefsExtractsCommands(t *testing.T) {
-	content := strings.Join([]string{
-		"运行 `lark-cli contact +search-user --query 张三` 搜索", // inline code
-		"```bash",
-		"lark-cli api GET /open-apis/x --params '{}'", // bash block
-		"```",
-		"用 lark-cli mail user_mailbox.messages batch_modify 即可", // bare prose command
-		"npx foo | lark-cli api GET /y",                         // after a pipe
-	}, "\n")
-	refs := parseRefs(content)
-	if len(refs) != 4 {
-		t.Fatalf("expected 4 refs, got %d: %v", len(refs), refWordsOf(refs))
-	}
-	if got := refs[0]; strings.Join(got.words, " ") != "contact +search-user" ||
-		len(got.flags) != 1 || got.flags[0] != "--query" {
-		t.Errorf("ref0 = %+v", got)
-	}
-	if got := refs[1]; strings.Join(got.words, " ") != "api GET /open-apis/x" {
-		t.Errorf("ref1 words = %v", got.words)
-	}
-}
-
-func TestCmdExampleParseRefsFiltersPlaceholdersAndProse(t *testing.T) {
-	// A line whose first word is prose yields no command at all.
-	if refs := parseRefs("lark-cli 就能搞定这件事"); len(refs) != 0 {
-		t.Errorf("prose-first line should yield 0 refs, got %v", refWordsOf(refs))
-	}
-	// Syntax templates / trailing prose may leave a real leading word ("mail"),
-	// but no placeholder or CJK token may leak into the command words — that is
-	// what prevents false positives like an "<resource>" unknown-command report.
-	for _, line := range []string{
-		"lark-cli mail <resource> <method> [flags]",
-		"lark-cli apps +<verb> [flags]",
-		"lark-cli base +...",
-		"lark-cli mail 写信场景下的格式说明",
-	} {
-		for _, r := range parseRefs(line) {
-			for _, w := range r.words {
-				if isPlaceholderOrProse(w) {
-					t.Errorf("%q: placeholder/prose token %q leaked into words %v", line, w, r.words)
-				}
-			}
-		}
-	}
-}
-
-func TestCmdExampleParseRefsStripsTrailingJunk(t *testing.T) {
-	// frontmatter-style quoted value: the trailing quote must not bleed into the flag
-	refs := parseRefs(`cliHelp: "lark-cli contact --help"`)
-	if len(refs) != 1 {
-		t.Fatalf("expected 1 ref, got %d", len(refs))
-	}
-	if len(refs[0].flags) != 1 || refs[0].flags[0] != "--help" {
-		t.Errorf("expected flag --help, got %v", refs[0].flags)
-	}
-	// bare "-" (stdin marker) and "=value" suffix
-	refs = parseRefs("lark-cli api GET /x --params={} --data -")
-	if len(refs) != 1 {
-		t.Fatalf("expected 1 ref, got %d", len(refs))
-	}
-	flags := strings.Join(refs[0].flags, " ")
-	if flags != "--params --data" {
-		t.Errorf("expected '--params --data', got %q", flags)
-	}
-}
-
-func TestCmdExampleCheck(t *testing.T) {
-	c := testCatalog()
-	tests := []struct {
-		name     string
-		r        ref
-		wantKind string // "" = no finding
-		wantPath string
-	}{
-		{"valid shortcut", ref{words: []string{"contact", "+search-user"}, flags: []string{"--query"}}, "", ""},
-		{"valid leaf positional", ref{words: []string{"api", "GET", "/x"}}, "", ""},
-		{"unknown top command", ref{words: []string{"nope"}}, unknownCommand, "nope"},
-		{"group leftover = unknown subcommand",
-			ref{words: []string{"mail", "user_mailbox.messages", "batch_modify_message"}},
-			unknownCommand, "mail user_mailbox.messages batch_modify_message"},
-		{"unknown flag", ref{words: []string{"contact", "+search-user"}, flags: []string{"--nope"}}, unknownFlag, "contact +search-user"},
-		{"universal flag ok", ref{words: []string{"contact", "+search-user"}, flags: []string{"--help"}}, "", ""},
-	}
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			fs := checkRefs(c, []ref{tt.r})
-			if tt.wantKind == "" {
-				if len(fs) != 0 {
-					t.Fatalf("expected no finding, got %+v", fs)
-				}
-				return
-			}
-			if len(fs) != 1 {
-				t.Fatalf("expected 1 finding, got %d: %+v", len(fs), fs)
-			}
-			if fs[0].kind != tt.wantKind || fs[0].path != tt.wantPath {
-				t.Errorf("got kind=%s path=%q, want kind=%s path=%q", fs[0].kind, fs[0].path, tt.wantKind, tt.wantPath)
-			}
-		})
-	}
-}
-
-func TestCmdExampleCheckSuggestsNearest(t *testing.T) {
-	c := testCatalog()
-	fs := checkRefs(c, []ref{{words: []string{"mail", "user_mailbox.messages", "batch_modify_message"}}})
-	if len(fs) != 1 || fs[0].suggest != "mail user_mailbox.messages batch_modify" {
-		t.Fatalf("expected suggestion 'mail user_mailbox.messages batch_modify', got %+v", fs)
-	}
-}
-
-// TestCmdExampleParseRefsRobustness covers the parser edge cases hardened after
-// review: backslash continuation, underscore flags, $(...) substitution, glued
-// separators, trailing punctuation, and the "..." placeholder.
-func TestCmdExampleParseRefsRobustness(t *testing.T) {
-	cases := []struct {
-		name, content, wantWords, wantFlags string
-		wantRefs                            int
-	}{
-		{"backslash continuation joins flags",
-			"lark-cli contact +search-user \\\n  --query foo \\\n  --as user",
-			"contact +search-user", "--query --as", 1},
-		{"underscore flag not truncated",
-			"lark-cli whiteboard +update --input_format mermaid",
-			"whiteboard +update", "--input_format", 1},
-		{"command-substitution flags ignored",
-			`lark-cli slides x create --data "$(jq -n --arg c '{}')" --as user`,
-			"slides x create", "--data --as", 1},
-		{"glued separator truncates",
-			"lark-cli auth login; echo done",
-			"auth login", "", 1},
-		{"trailing CJK punctuation stripped",
-			"用 lark-cli auth login。",
-			"auth login", "", 1},
-		{"ellipsis placeholder stays placeholder",
-			"lark-cli base +...",
-			"base", "", 1},
-	}
-	for _, tt := range cases {
-		t.Run(tt.name, func(t *testing.T) {
-			refs := parseRefs(tt.content)
-			if len(refs) != tt.wantRefs {
-				t.Fatalf("refs=%d want %d: %v", len(refs), tt.wantRefs, refWordsOf(refs))
-			}
-			if tt.wantRefs == 0 {
-				return
-			}
-			if got := strings.Join(refs[0].words, " "); got != tt.wantWords {
-				t.Errorf("words=%q want %q", got, tt.wantWords)
-			}
-			if got := strings.Join(refs[0].flags, " "); got != tt.wantFlags {
-				t.Errorf("flags=%q want %q", got, tt.wantFlags)
-			}
-		})
-	}
-}
--- a/cmd/command_catalog_path_test.go
+++ b/cmd/command_catalog_path_test.go
@@ -1,52 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd
-
-import (
-	"reflect"
-	"testing"
-
-	"github.com/spf13/cobra"
-)
-
-// TestCommandCatalogPath pins that the auth-hint path reconstruction inverts the
-// service command tree for any depth — flat dotted resources AND genuinely
-// nested resources — so it round-trips through apicatalog.Resolve instead of
-// assuming a fixed root->service->resource->method shape.
-func TestCommandCatalogPath(t *testing.T) {
-	chain := func(names ...string) *cobra.Command {
-		var parent, leaf *cobra.Command
-		for _, n := range names {
-			c := &cobra.Command{Use: n}
-			if parent != nil {
-				parent.AddCommand(c)
-			}
-			parent = c
-			leaf = c
-		}
-		return leaf
-	}
-
-	tests := []struct {
-		name string
-		leaf *cobra.Command
-		want []string
-	}{
-		{"flat dotted resource", chain("lark-cli", "im", "chat.members", "create"), []string{"im", "chat.members", "create"}},
-		{"nested resources", chain("lark-cli", "im", "spaces", "items", "get"), []string{"im", "spaces", "items", "get"}},
-		{"service level", chain("lark-cli", "im"), []string{"im"}},
-	}
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			if got := commandCatalogPath(tt.leaf); !reflect.DeepEqual(got, tt.want) {
-				t.Errorf("commandCatalogPath = %v, want %v", got, tt.want)
-			}
-		})
-	}
-
-	// The root command (no parent) has no catalog path.
-	if got := commandCatalogPath(&cobra.Command{Use: "lark-cli"}); len(got) != 0 {
-		t.Errorf("root path = %v, want empty", got)
-	}
-}
--- a/cmd/completion/completion.go
+++ b/cmd/completion/completion.go
@@ -4,7 +4,8 @@
 package completion

 import (
-	"github.com/larksuite/cli/errs"
+	"fmt"
+
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/spf13/cobra"
 )
@@ -31,9 +32,7 @@ func NewCmdCompletion(f *cmdutil.Factory) *cobra.Command {
 			case "powershell":
 				return root.GenPowerShellCompletionWithDesc(out)
 			default:
-				return errs.NewValidationError(errs.SubtypeInvalidArgument,
-					"unsupported shell: %s", args[0]).
-					WithHint("supported shells: bash, zsh, fish, powershell")
+				return fmt.Errorf("unsupported shell: %s", args[0])
 			}
 		},
 	}
--- a/cmd/config/bind.go
+++ b/cmd/config/bind.go
@@ -212,7 +212,10 @@ func finalizeSource(opts *BindOptions) (string, error) {
 	if opts.IsTUI && !opts.langExplicit {
 		lang, err := promptLangSelection()
 		if err != nil {
-			return "", langSelectionError(err)
+			if err == huh.ErrUserAborted {
+				return "", output.ErrBare(1)
+			}
+			return "", output.Errorf(output.ExitInternal, "internal", "language selection failed: %v", err)
 		}
 		opts.Lang = string(lang)
 		opts.UILang = lang
--- a/cmd/config/bind_test.go
+++ b/cmd/config/bind_test.go
@@ -20,29 +20,35 @@ import (
 	"github.com/larksuite/cli/internal/output"
 )

-// wantErrDetail is the normalized comparison shape for a typed error's wire
-// fields: Type is the error's Category string ("validation", "config", ...),
-// alongside Message and Hint.
-type wantErrDetail struct {
-	Type    string
-	Message string
-	Hint    string
-}
-
-// assertExitError checks the full structured error in one assertion against a
-// typed error (ValidationError or ConfigError), normalizing its Category /
-// Message / Hint to wantDetail.
-func assertExitError(t *testing.T, err error, wantCode int, wantDetail wantErrDetail) {
+// assertExitError checks the full structured error in one assertion. It
+// accepts both *output.ExitError (used by output.ErrWithHint) and the
+// typed errors (ValidationError, ConfigError) — they normalize to the same
+// wantDetail fields. The wantDetail.Type is matched against the typed error's
+// Category string ("validation", "config", etc.).
+func assertExitError(t *testing.T, err error, wantCode int, wantDetail output.ErrDetail) {
 	t.Helper()
 	if err == nil {
 		t.Fatal("expected error, got nil")
 	}
+	var exitErr *output.ExitError
+	if errors.As(err, &exitErr) {
+		if exitErr.Code != wantCode {
+			t.Errorf("exit code = %d, want %d", exitErr.Code, wantCode)
+		}
+		if exitErr.Detail == nil {
+			t.Fatal("expected non-nil error detail")
+		}
+		if !reflect.DeepEqual(*exitErr.Detail, wantDetail) {
+			t.Errorf("error detail mismatch:\n  got:  %+v\n  want: %+v", *exitErr.Detail, wantDetail)
+		}
+		return
+	}
 	var ve *errs.ValidationError
 	if errors.As(err, &ve) {
 		if got := output.ExitCodeOf(err); got != wantCode {
 			t.Errorf("exit code = %d, want %d", got, wantCode)
 		}
-		gotDetail := wantErrDetail{Type: string(ve.Category), Message: ve.Message, Hint: ve.Hint}
+		gotDetail := output.ErrDetail{Type: string(ve.Category), Message: ve.Message, Hint: ve.Hint}
 		if !reflect.DeepEqual(gotDetail, wantDetail) {
 			t.Errorf("validation error mismatch:\n  got:  %+v\n  want: %+v", gotDetail, wantDetail)
 		}
@@ -53,13 +59,13 @@ func assertExitError(t *testing.T, err error, wantCode int, wantDetail wantErrDe
 		if got := output.ExitCodeOf(err); got != wantCode {
 			t.Errorf("exit code = %d, want %d", got, wantCode)
 		}
-		gotDetail := wantErrDetail{Type: string(ce.Category), Message: ce.Message, Hint: ce.Hint}
+		gotDetail := output.ErrDetail{Type: string(ce.Category), Message: ce.Message, Hint: ce.Hint}
 		if !reflect.DeepEqual(gotDetail, wantDetail) {
 			t.Errorf("config error mismatch:\n  got:  %+v\n  want: %+v", gotDetail, wantDetail)
 		}
 		return
 	}
-	t.Fatalf("error type = %T, want *errs.ValidationError / *errs.ConfigError; error = %v", err, err)
+	t.Fatalf("error type = %T, want *output.ExitError or *errs.ValidationError / *errs.ConfigError; error = %v", err, err)
 }

 // assertEnvelope decodes stdout and checks it matches want exactly — every key
@@ -173,21 +179,15 @@ func TestConfigBindRun_InvalidLang(t *testing.T) {
 			if err == nil {
 				t.Fatalf("expected validation error for --lang %q, got nil", tc.lang)
 			}
-			var valErr *errs.ValidationError
-			if !errors.As(err, &valErr) {
-				t.Fatalf("expected *errs.ValidationError, got %T: %v", err, err)
+			exitErr, ok := err.(*output.ExitError)
+			if !ok {
+				t.Fatalf("expected *output.ExitError, got %T: %v", err, err)
 			}
-			if valErr.Subtype != errs.SubtypeInvalidArgument {
-				t.Errorf("subtype = %q, want %q", valErr.Subtype, errs.SubtypeInvalidArgument)
+			if exitErr.Code != output.ExitValidation {
+				t.Errorf("exit code = %d, want %d (validation)", exitErr.Code, output.ExitValidation)
 			}
-			if valErr.Param != "--lang" {
-				t.Errorf("param = %q, want %q", valErr.Param, "--lang")
-			}
-			if got := output.ExitCodeOf(err); got != output.ExitValidation {
-				t.Errorf("exit code = %d, want %d (validation)", got, output.ExitValidation)
-			}
-			if !strings.Contains(err.Error(), "invalid --lang") {
-				t.Errorf("error message %q does not contain 'invalid --lang'", err.Error())
+			if !strings.Contains(exitErr.Error(), "invalid --lang") {
+				t.Errorf("error message %q does not contain 'invalid --lang'", exitErr.Error())
 			}
 		})
 	}
@@ -365,7 +365,7 @@ func TestConfigBindRun_InvalidSource(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "invalid"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `invalid --source "invalid"; valid values: openclaw, hermes, lark-channel`,
 	})
@@ -382,7 +382,7 @@ func TestConfigBindRun_MissingSourceNonTTY(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	// TestFactory has IsTerminal=false by default
 	err := configBindRun(&BindOptions{Factory: f, Source: ""})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: "cannot determine Agent source: no --source flag and no Agent environment detected",
 		Hint:    "pass --source openclaw|hermes|lark-channel, or run this command inside the corresponding Agent context",
@@ -421,7 +421,7 @@ func TestConfigBindRun_SourceEnvMismatch_OpenClawFlagInHermesEnv(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--source "openclaw" does not match detected Agent environment (hermes)`,
 		Hint:    "remove --source to auto-detect, or run this command in the correct Agent context",
@@ -437,7 +437,7 @@ func TestConfigBindRun_SourceEnvMismatch_HermesFlagInOpenClawEnv(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "hermes"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--source "hermes" does not match detected Agent environment (openclaw)`,
 		Hint:    "remove --source to auto-detect, or run this command in the correct Agent context",
@@ -566,7 +566,7 @@ func TestConfigBindRun_HermesMissingEnvFile(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "hermes"})
 	envPath := filepath.Join(hermesHome, ".env")
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "failed to read Hermes config: open " + envPath + ": no such file or directory",
 		Hint:    "verify Hermes is installed and configured at " + envPath,
@@ -584,7 +584,7 @@ func TestConfigBindRun_OpenClawMissingFile(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw"})
 	configPath := filepath.Join(openclawHome, ".openclaw", "openclaw.json")
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "cannot read " + configPath + ": open " + configPath + ": no such file or directory",
 		Hint:    "verify OpenClaw is installed and configured",
@@ -731,7 +731,7 @@ func TestConfigBindRun_SourceEnvMismatch_LarkChannelFlagInOpenClawEnv(t *testing

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "lark-channel"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--source "lark-channel" does not match detected Agent environment (openclaw)`,
 		Hint:    "remove --source to auto-detect, or run this command in the correct Agent context",
@@ -750,7 +750,7 @@ func TestConfigBindRun_LarkChannelMissingFile(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "lark-channel"})
 	configPath := filepath.Join(fakeHome, ".lark-channel", "config.json")
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "cannot read " + configPath + ": open " + configPath + ": no such file or directory",
 		Hint:    "verify lark-channel-bridge is installed and configured",
@@ -770,7 +770,7 @@ func TestConfigBindRun_LarkChannelEmptyAppID(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "lark-channel"})
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "accounts.app.id missing in " + configPath,
 		Hint:    "run lark-channel-bridge's setup to populate the app credential",
@@ -789,7 +789,7 @@ func TestConfigBindRun_LarkChannelEmptySecret(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "lark-channel"})
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "accounts.app.secret is empty in " + configPath,
 		Hint:    "run lark-channel-bridge's setup to populate the app credential",
@@ -835,19 +835,17 @@ func TestConfigShowRun_AgentWorkspaceNotBound(t *testing.T) {
 		t.Fatal("expected error for unbound workspace")
 	}
 	// Should be a structured ConfigError suggesting config bind, not config init.
-	var cfgErr *errs.ConfigError
+	var cfgErr *core.ConfigError
 	if !errors.As(err, &cfgErr) {
-		t.Fatalf("error type = %T, want *errs.ConfigError", err)
+		t.Fatalf("error type = %T, want *core.ConfigError", err)
 	}
 	// Config errors share ExitAuth (3); the workspace is detected but no
 	// binding exists yet, which is a config error.
-	if got := output.ExitCodeOf(err); got != output.ExitAuth {
-		t.Errorf("exit code = %d, want %d (config category → ExitAuth)", got, output.ExitAuth)
+	if cfgErr.Code != output.ExitAuth {
+		t.Errorf("exit code = %d, want %d (config category → ExitAuth)", cfgErr.Code, output.ExitAuth)
 	}
-	// The workspace name stays out of the wire subtype; it only appears in
-	// the message.
-	if cfgErr.Subtype != errs.SubtypeNotConfigured {
-		t.Errorf("subtype = %q, want not_configured", cfgErr.Subtype)
+	if cfgErr.Type != "openclaw" {
+		t.Errorf("type = %q, want %q", cfgErr.Type, "openclaw")
 	}
 	if !strings.Contains(cfgErr.Message, "openclaw context detected") {
 		t.Errorf("message missing 'openclaw context detected': %q", cfgErr.Message)
@@ -1189,7 +1187,7 @@ func TestConfigBindRun_OpenClawMultiAccount_TTYFlagMode(t *testing.T) {
 	// iterates a map — ordering is non-deterministic. DeepEqual inline against
 	// each accepted variant so every ErrDetail field (Type, Code, Message,
 	// Hint, ConsoleURL, Detail, and any future addition) is still compared.
-	base := wantErrDetail{
+	base := output.ErrDetail{
 		Type:    "validation",
 		Message: "multiple accounts in openclaw.json; pass --app-id <id>",
 	}
@@ -1205,7 +1203,7 @@ func TestConfigBindRun_OpenClawMultiAccount_TTYFlagMode(t *testing.T) {
 	if !errors.As(err, &ve) {
 		t.Fatalf("error type = %T, want *errs.ValidationError; err = %v", err, err)
 	}
-	got := wantErrDetail{Type: string(ve.Category), Message: ve.Message, Hint: ve.Hint}
+	got := output.ErrDetail{Type: string(ve.Category), Message: ve.Message, Hint: ve.Hint}
 	if !reflect.DeepEqual(got, wantWorkFirst) && !reflect.DeepEqual(got, wantPersonalFirst) {
 		t.Errorf("error detail did not match any accepted variant:\n  got:  %+v\n  want: %+v OR %+v",
 			got, wantWorkFirst, wantPersonalFirst)
@@ -1232,7 +1230,7 @@ func TestConfigBindRun_OpenClawMultiAccount_WrongAppID(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw", AppID: "nonexistent"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--app-id "nonexistent" not found in openclaw.json`,
 		Hint:    "available app IDs:\n  cli_only_one",
@@ -1252,7 +1250,7 @@ func TestConfigBindRun_InvalidIdentity(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "hermes", Identity: "invalid"})
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `invalid --identity "invalid"; valid values: bot-only, user-default`,
 	})
@@ -1538,7 +1536,7 @@ func TestConfigBindRun_HermesMissingAppID(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "hermes"})
 	envPath := filepath.Join(hermesHome, ".env")
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "FEISHU_APP_ID not found in " + envPath,
 		Hint:    "run 'hermes setup' to configure Feishu credentials",
@@ -1558,7 +1556,7 @@ func TestConfigBindRun_HermesMissingAppSecret(t *testing.T) {
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "hermes"})
 	envPath := filepath.Join(hermesHome, ".env")
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "FEISHU_APP_SECRET not found in " + envPath,
 		Hint:    "run 'hermes setup' to configure Feishu credentials",
@@ -1584,7 +1582,7 @@ func TestConfigBindRun_OpenClawMissingFeishu(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw"})
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "openclaw.json missing channels.feishu section",
 		Hint:    "configure Feishu in OpenClaw first",
@@ -1612,7 +1610,7 @@ func TestConfigBindRun_OpenClawEmptyAppSecret(t *testing.T) {
 	openclawPath := filepath.Join(openclawDir, "openclaw.json")
 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw"})
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "appSecret is empty for app cli_no_secret in " + openclawPath,
 		Hint:    "configure channels.feishu.appSecret in openclaw.json",
@@ -1674,7 +1672,7 @@ func TestConfigBindRun_OpenClawDisabledAccount(t *testing.T) {

 	f, _, _, _ := cmdutil.TestFactory(t, nil)
 	err := configBindRun(&BindOptions{Factory: f, Source: "openclaw"})
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "no Feishu app configured in openclaw.json",
 		Hint:    "configure channels.feishu.appId in openclaw.json",
--- a/cmd/config/binder_test.go
+++ b/cmd/config/binder_test.go
@@ -51,7 +51,7 @@ func assertCandidate(t *testing.T, got *Candidate, want Candidate) {
 func TestSelectCandidate_ZeroCandidates_OpenClaw(t *testing.T) {
 	b := &fakeBinder{name: "openclaw", path: "/tmp/openclaw.json"}
 	_, err := selectCandidate(b, nil, "", false, tuiUnreachable(t))
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "no Feishu app configured in openclaw.json",
 		Hint:    "configure channels.feishu.appId in openclaw.json",
@@ -64,7 +64,7 @@ func TestSelectCandidate_ZeroCandidates_GenericSource(t *testing.T) {
 	// even before it has a bespoke error message.
 	b := &fakeBinder{name: "hermes", path: "/tmp/.env"}
 	_, err := selectCandidate(b, nil, "", false, tuiUnreachable(t))
-	assertExitError(t, err, output.ExitAuth, wantErrDetail{
+	assertExitError(t, err, output.ExitAuth, output.ErrDetail{
 		Type:    "config",
 		Message: "hermes: no app configured",
 	})
@@ -100,7 +100,7 @@ func TestSelectCandidate_AppIDFlag_NoMatch(t *testing.T) {
 		{AppID: "cli_home", Label: "home"},
 	}
 	_, err := selectCandidate(b, candidates, "nonexistent", false, tuiUnreachable(t))
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--app-id "nonexistent" not found in openclaw.json`,
 		Hint:    "available app IDs:\n  cli_work (work)\n  cli_home (home)",
@@ -117,7 +117,7 @@ func TestSelectCandidate_MultiCandidate_NoFlag_NonTUI(t *testing.T) {
 		{AppID: "cli_home", Label: "home"},
 	}
 	_, err := selectCandidate(b, candidates, "", false, tuiUnreachable(t))
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: "multiple accounts in openclaw.json; pass --app-id <id>",
 		Hint:    "available app IDs:\n  cli_work (work)\n  cli_home (home)",
@@ -152,7 +152,7 @@ func TestSelectCandidate_SingleCandidate_WrongFlag(t *testing.T) {
 	b := &fakeBinder{name: "openclaw", path: "/tmp/openclaw.json"}
 	candidates := []Candidate{{AppID: "cli_only"}}
 	_, err := selectCandidate(b, candidates, "nonexistent", false, tuiUnreachable(t))
-	assertExitError(t, err, output.ExitValidation, wantErrDetail{
+	assertExitError(t, err, output.ExitValidation, output.ErrDetail{
 		Type:    "validation",
 		Message: `--app-id "nonexistent" not found in openclaw.json`,
 		Hint:    "available app IDs:\n  cli_only",
--- a/cmd/config/config_test.go
+++ b/cmd/config/config_test.go
@@ -12,7 +12,6 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
 	extcred "github.com/larksuite/cli/extension/credential"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -93,16 +92,16 @@ func TestConfigShowRun_NotConfiguredReturnsStructuredError(t *testing.T) {
 		t.Fatal("expected error")
 	}

-	var cfgErr *errs.ConfigError
+	var cfgErr *core.ConfigError
 	if !errors.As(err, &cfgErr) {
-		t.Fatalf("error type = %T, want *errs.ConfigError", err)
+		t.Fatalf("error type = %T, want *core.ConfigError", err)
 	}
 	// Config errors share ExitAuth (3), not ExitValidation.
-	if got := output.ExitCodeOf(err); got != output.ExitAuth {
-		t.Fatalf("exit code = %d, want %d (config category → ExitAuth)", got, output.ExitAuth)
+	if cfgErr.Code != output.ExitAuth {
+		t.Fatalf("exit code = %d, want %d (config category → ExitAuth)", cfgErr.Code, output.ExitAuth)
 	}
-	if cfgErr.Subtype != errs.SubtypeNotConfigured || cfgErr.Message != "not configured" {
-		t.Fatalf("detail = %+v, want not_configured/not configured", cfgErr)
+	if cfgErr.Type != "config" || cfgErr.Message != "not configured" {
+		t.Fatalf("detail = %+v, want config/not configured", cfgErr)
 	}
 }

@@ -234,21 +233,15 @@ func TestConfigInitCmd_InvalidLang(t *testing.T) {
 			if err == nil {
 				t.Fatalf("expected validation error for --lang %q, got nil", tc.lang)
 			}
-			var valErr *errs.ValidationError
-			if !errors.As(err, &valErr) {
-				t.Fatalf("expected *errs.ValidationError, got %T: %v", err, err)
+			exitErr, ok := err.(*output.ExitError)
+			if !ok {
+				t.Fatalf("expected *output.ExitError, got %T: %v", err, err)
 			}
-			if valErr.Subtype != errs.SubtypeInvalidArgument {
-				t.Errorf("subtype = %q, want %q", valErr.Subtype, errs.SubtypeInvalidArgument)
+			if exitErr.Code != output.ExitValidation {
+				t.Errorf("exit code = %d, want %d (validation)", exitErr.Code, output.ExitValidation)
 			}
-			if valErr.Param != "--lang" {
-				t.Errorf("param = %q, want %q", valErr.Param, "--lang")
-			}
-			if got := output.ExitCodeOf(err); got != output.ExitValidation {
-				t.Errorf("exit code = %d, want %d (validation)", got, output.ExitValidation)
-			}
-			if !strings.Contains(err.Error(), "invalid --lang") {
-				t.Errorf("error message %q does not contain 'invalid --lang'", err.Error())
+			if !strings.Contains(exitErr.Error(), "invalid --lang") {
+				t.Errorf("error message %q does not contain 'invalid --lang'", exitErr.Error())
 			}
 		})
 	}
@@ -392,38 +385,8 @@ func TestSaveAsProfile_RejectsProfileNameCollisionWithExistingAppID(t *testing.T
 	if err == nil {
 		t.Fatal("expected conflict error")
 	}
-	// A name/appId conflict is user input — a typed validation error naming the
-	// offending flag, not a system storage failure.
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("error type = %T, want *errs.ValidationError; err=%v", err, err)
-	}
-	if verr.Subtype != errs.SubtypeInvalidArgument {
-		t.Errorf("subtype = %q, want invalid_argument", verr.Subtype)
-	}
-	if verr.Param != "--name" {
-		t.Errorf("param = %q, want --name", verr.Param)
-	}
-	if output.ExitCodeOf(err) != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (validation)", output.ExitCodeOf(err), output.ExitValidation)
-	}
-	if !strings.Contains(verr.Message, "conflicts with existing appId") {
-		t.Errorf("message = %q, want conflict description", verr.Message)
-	}
-}
-
-// TestWrapSaveConfigError_PassesTypedValidationThrough pins that a user-input
-// validation error (e.g. the --name conflict) is not reclassified as an
-// internal storage failure on its way up through the save call sites.
-func TestWrapSaveConfigError_PassesTypedValidationThrough(t *testing.T) {
-	conflict := errs.NewValidationError(errs.SubtypeInvalidArgument, "name conflict").WithParam("--name")
-	var verr *errs.ValidationError
-	if !errors.As(wrapSaveConfigError(conflict), &verr) {
-		t.Fatalf("typed validation must pass through unchanged, got %T", wrapSaveConfigError(conflict))
-	}
-	var ierr *errs.InternalError
-	if !errors.As(wrapSaveConfigError(errors.New("disk full")), &ierr) || ierr.Subtype != errs.SubtypeStorage {
-		t.Fatalf("untyped failure must become internal/storage")
+	if !strings.Contains(err.Error(), "conflicts with existing appId") {
+		t.Fatalf("error = %v, want conflict with existing appId", err)
 	}
 }

--- a/cmd/config/init.go
+++ b/cmd/config/init.go
@@ -6,11 +6,13 @@ package config
 import (
 	"bufio"
 	"context"
+	"errors"
 	"fmt"
 	"io"
 	"os"
 	"strings"

+	"github.com/charmbracelet/huh"
 	"github.com/spf13/cobra"

 	"github.com/larksuite/cli/errs"
@@ -125,9 +127,12 @@ func guardAgentWorkspace(opts *ConfigInitOptions) error {
 	if ws.IsLocal() {
 		return nil
 	}
-	return errs.NewConfigError(errs.SubtypeNotConfigured,
-		"config init is refused inside %s context (would create a parallel app and shadow the existing %s binding)", ws.Display(), ws.Display()).
-		WithHint("see `lark-cli config bind --help` to bind lark-cli to the Agent's existing app instead. Pass --force-init only if the user explicitly wants a separate app in this workspace.")
+	return &core.ConfigError{
+		Code:    2,
+		Type:    ws.Display(),
+		Message: fmt.Sprintf("config init is refused inside %s context (would create a parallel app and shadow the existing %s binding)", ws.Display(), ws.Display()),
+		Hint:    "see `lark-cli config bind --help` to bind lark-cli to the Agent's existing app instead. Pass --force-init only if the user explicitly wants a separate app in this workspace.",
+	}
 }

 // hasAnyNonInteractiveFlag returns true if any non-interactive flag is set.
@@ -178,20 +183,6 @@ func saveInitConfig(profileName string, existing *core.MultiAppConfig, f *cmduti
 	return saveAsOnlyApp(appId, secret, brand, string(preferredLang(i18n.Lang(lang), prior)))
 }

-// wrapSaveConfigError passes an already-typed error (e.g. the --name conflict
-// validation error from saveAsProfile) through unchanged, and classifies any
-// other failure as an internal storage error. Without the passthrough a user
-// input error would surface to agents as a system storage failure.
-func wrapSaveConfigError(err error) error {
-	if err == nil {
-		return nil
-	}
-	if _, ok := errs.ProblemOf(err); ok {
-		return err
-	}
-	return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
-}
-
 // saveAsProfile appends or updates a named profile in the config.
 // If a profile with the same name exists, it updates it; otherwise appends.
 // When updating, cleans up old keychain secrets if AppId changed.
@@ -216,9 +207,7 @@ func saveAsProfile(existing *core.MultiAppConfig, kc keychain.KeychainAccess, pr
 		multi.Apps[idx].Lang = preferredLang(i18n.Lang(lang), multi.Apps[idx].Lang)
 	} else {
 		if findAppIndexByAppID(multi, profileName) >= 0 {
-			return errs.NewValidationError(errs.SubtypeInvalidArgument,
-				"profile name %q conflicts with existing appId", profileName).
-				WithParam("--name")
+			return fmt.Errorf("profile name %q conflicts with existing appId", profileName)
 		}
 		// Append new profile
 		multi.Apps = append(multi.Apps, core.AppConfig{
@@ -260,8 +249,8 @@ func findAppIndexByAppID(multi *core.MultiAppConfig, appID string) int {
 // wrapUpdateExistingProfileErr classifies the error returned by
 // updateExistingProfileWithoutSecret. Typed errors (e.g. *errs.ValidationError
 // for blank-input) pass through unchanged so their exit code semantics
-// survive; everything else (filesystem, keychain, etc.) is wrapped as
-// InternalError.
+// survive; legacy *output.ExitError also passes through; everything else
+// (filesystem, keychain, etc.) is wrapped as InternalError.
 func wrapUpdateExistingProfileErr(err error) error {
 	if err == nil {
 		return nil
@@ -269,6 +258,10 @@ func wrapUpdateExistingProfileErr(err error) error {
 	if errs.IsTyped(err) {
 		return err
 	}
+	var exitErr *output.ExitError
+	if errors.As(err, &exitErr) {
+		return err
+	}
 	return errs.NewInternalError(errs.SubtypeSDKError, "failed to save config: %v", err).WithCause(err)
 }

@@ -343,7 +336,7 @@ func configInitRun(opts *ConfigInitOptions) error {
 			return errs.NewInternalError(errs.SubtypeSDKError, "%v", err).WithCause(err)
 		}
 		if err := saveInitConfig(opts.ProfileName, existing, f, opts.AppID, secret, brand, opts.Lang); err != nil {
-			return wrapSaveConfigError(err)
+			return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
 		}
 		output.PrintSuccess(f.IOStreams.ErrOut, fmt.Sprintf("Configuration saved to %s", core.GetConfigPath()))
 		printLangPreferenceConfirmation(opts)
@@ -360,7 +353,10 @@ func configInitRun(opts *ConfigInitOptions) error {
 	if f.IOStreams.IsTerminal && !opts.langExplicit && !opts.hasAnyNonInteractiveFlag() {
 		lang, err := promptLangSelection()
 		if err != nil {
-			return langSelectionError(err)
+			if err == huh.ErrUserAborted {
+				return output.ErrBare(1)
+			}
+			return output.Errorf(output.ExitInternal, "internal", "language selection failed: %v", err)
 		}
 		opts.Lang = string(lang)
 		opts.UILang = lang
@@ -383,7 +379,7 @@ func configInitRun(opts *ConfigInitOptions) error {
 			return errs.NewInternalError(errs.SubtypeSDKError, "%v", err).WithCause(err)
 		}
 		if err := saveInitConfig(opts.ProfileName, existing, f, result.AppID, secret, result.Brand, opts.Lang); err != nil {
-			return wrapSaveConfigError(err)
+			return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
 		}
 		printLangPreferenceConfirmation(opts)
 		output.PrintJson(f.IOStreams.Out, map[string]interface{}{"appId": result.AppID, "appSecret": "****", "brand": result.Brand})
@@ -413,7 +409,7 @@ func configInitRun(opts *ConfigInitOptions) error {
 				return errs.NewInternalError(errs.SubtypeSDKError, "%v", err).WithCause(err)
 			}
 			if err := saveInitConfig(opts.ProfileName, existing, f, result.AppID, secret, result.Brand, opts.Lang); err != nil {
-				return wrapSaveConfigError(err)
+				return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
 			}
 		} else if result.Mode == "existing" && result.AppID != "" {
 			// Existing app with unchanged secret — update app ID and brand only
@@ -518,7 +514,7 @@ func configInitRun(opts *ConfigInitOptions) error {
 		return errs.NewInternalError(errs.SubtypeSDKError, "%v", err).WithCause(err)
 	}
 	if err := saveInitConfig(opts.ProfileName, existing, f, resolvedAppId, storedSecret, parseBrand(resolvedBrand), opts.Lang); err != nil {
-		return wrapSaveConfigError(err)
+		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
 	}
 	output.PrintSuccess(f.IOStreams.ErrOut, fmt.Sprintf("Configuration saved to %s", core.GetConfigPath()))
 	printLangPreferenceConfirmation(opts)
--- a/cmd/config/init_guard_test.go
+++ b/cmd/config/init_guard_test.go
@@ -8,7 +8,7 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
+	"github.com/larksuite/cli/internal/core"
 )

 func TestGuardAgentWorkspace_LocalAllows(t *testing.T) {
@@ -26,15 +26,12 @@ func TestGuardAgentWorkspace_OpenClawRefuses(t *testing.T) {
 	if err == nil {
 		t.Fatal("expected refusal in OpenClaw context, got nil")
 	}
-	var cfgErr *errs.ConfigError
+	var cfgErr *core.ConfigError
 	if !errors.As(err, &cfgErr) {
-		t.Fatalf("error type = %T, want *errs.ConfigError", err)
+		t.Fatalf("error type = %T, want *core.ConfigError", err)
 	}
-	if cfgErr.Subtype != errs.SubtypeNotConfigured {
-		t.Errorf("subtype = %q, want not_configured", cfgErr.Subtype)
-	}
-	if !strings.Contains(cfgErr.Message, "openclaw") {
-		t.Errorf("message must name the openclaw workspace; got %q", cfgErr.Message)
+	if cfgErr.Type != "openclaw" {
+		t.Errorf("type = %q, want %q", cfgErr.Type, "openclaw")
 	}
 	if !strings.Contains(cfgErr.Hint, "config bind --help") {
 		t.Errorf("hint must point to config bind --help; got %q", cfgErr.Hint)
@@ -51,15 +48,12 @@ func TestGuardAgentWorkspace_HermesRefuses(t *testing.T) {
 	if err == nil {
 		t.Fatal("expected refusal in Hermes context, got nil")
 	}
-	var cfgErr *errs.ConfigError
+	var cfgErr *core.ConfigError
 	if !errors.As(err, &cfgErr) {
-		t.Fatalf("error type = %T, want *errs.ConfigError", err)
+		t.Fatalf("error type = %T, want *core.ConfigError", err)
 	}
-	if cfgErr.Subtype != errs.SubtypeNotConfigured {
-		t.Errorf("subtype = %q, want not_configured", cfgErr.Subtype)
-	}
-	if !strings.Contains(cfgErr.Message, "hermes") {
-		t.Errorf("message must name the hermes workspace; got %q", cfgErr.Message)
+	if cfgErr.Type != "hermes" {
+		t.Errorf("type = %q, want %q", cfgErr.Type, "hermes")
 	}
 }

--- a/cmd/config/init_messages.go
+++ b/cmd/config/init_messages.go
@@ -4,14 +4,10 @@
 package config

 import (
-	"errors"
-
 	"github.com/charmbracelet/huh"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/i18n"
-	"github.com/larksuite/cli/internal/output"
 )

 type initMsg struct {
@@ -101,12 +97,3 @@ func promptLangSelection() (i18n.Lang, error) {
 	}
 	return lang, nil
 }
-
-// langSelectionError maps a promptLangSelection failure to its exit surface:
-// user abort exits bare with code 1; any other failure is internal.
-func langSelectionError(err error) error {
-	if errors.Is(err, huh.ErrUserAborted) {
-		return output.ErrBare(1)
-	}
-	return errs.NewInternalError(errs.SubtypeUnknown, "language selection failed: %v", err).WithCause(err)
-}
--- a/cmd/config/init_probe.go
+++ b/cmd/config/init_probe.go
@@ -33,16 +33,15 @@ const probeTimeout = 3 * time.Second
 //
 //  1. A TAT request using the just-saved credentials. credential.FetchTAT
 //     returns a typed errs.* error (via the shared classifyTATResponseCode)
-//     only when the unified Token Endpoint deterministically rejected the
-//     credentials — an OAuth2 invalid_client / unauthorized_client classified as
-//     CategoryConfig / SubtypeInvalidClient, or whatever codemeta maps. That
-//     typed error is propagated so the root dispatcher renders the canonical
-//     envelope and `config init` exits non-zero — identical to how every other
-//     token-resolving command reports the same bad credentials. Ambiguous
-//     failures (transport errors, transient 5xx/server_error, JSON parse errors,
-//     timeouts) come back as raw untyped errors and are swallowed (return nil),
-//     so valid configurations are never disturbed by upstream noise.
-//     errs.IsTyped is the discriminator.
+//     only when the server deterministically rejected the credentials — a
+//     non-zero TAT body code, classified as CategoryConfig / SubtypeInvalidClient
+//     (10003 / 10014) or whatever codemeta maps. That typed error is propagated
+//     so the root dispatcher renders the canonical envelope and `config init`
+//     exits non-zero — identical to how every other token-resolving command
+//     reports the same bad credentials. Ambiguous failures (transport errors,
+//     HTTP non-200, JSON parse errors, timeouts) come back as raw untyped
+//     errors and are swallowed (return nil), so valid configurations are never
+//     disturbed by upstream noise. errs.IsTyped is the discriminator.
 //
 //  2. If TAT succeeded, a POST to the probe endpoint is fired. The outcome of
 //     that call (success, server error, timeout, parse failure) is always
--- a/cmd/config/init_probe_test.go
+++ b/cmd/config/init_probe_test.go
@@ -31,10 +31,10 @@ type fakeRT struct {

 func (f *fakeRT) RoundTrip(req *http.Request) (*http.Response, error) {
 	switch {
-	case strings.HasSuffix(req.URL.Path, "/oauth/v3/token"):
+	case strings.HasSuffix(req.URL.Path, "/auth/v3/tenant_access_token/internal"):
 		f.tatCalls++
 		if f.tatHandler == nil {
-			return jsonResp(200, `{"code":0,"access_token":"t-ok","token_type":"Bearer"}`), nil
+			return jsonResp(200, `{"code":0,"tenant_access_token":"t-ok"}`), nil
 		}
 		return f.tatHandler(req)
 	case strings.HasSuffix(req.URL.Path, "/application/v6/larksuite_cli_app/probe"):
@@ -84,15 +84,14 @@ func fakeFactory(t *testing.T, rt http.RoundTripper) (*cmdutil.Factory, *bytes.B
 }

 // assertConfigRejection asserts runProbe propagated a deterministic credential
-// rejection: a *errs.ConfigError (CategoryConfig / SubtypeInvalidClient). This
-// is the same typed error every other token-resolving command returns for the
-// same bad credentials, and nothing is written to stderr (the root dispatcher
-// renders the envelope). The numeric code is not asserted: the unified v3 Token
-// Endpoint reports invalid_client via the OAuth2 error string, not a Lark code.
-func assertConfigRejection(t *testing.T, err error, errBuf *bytes.Buffer) {
+// rejection: a *errs.ConfigError (CategoryConfig / SubtypeInvalidClient) with
+// the expected upstream code. This is the same typed error every other
+// token-resolving command returns for the same bad credentials, and nothing is
+// written to stderr (the root dispatcher renders the envelope).
+func assertConfigRejection(t *testing.T, err error, errBuf *bytes.Buffer, wantCode int) {
 	t.Helper()
 	if err == nil {
-		t.Fatal("expected *errs.ConfigError, got nil")
+		t.Fatalf("expected *errs.ConfigError (code %d), got nil", wantCode)
 	}
 	var cfgErr *errs.ConfigError
 	if !errors.As(err, &cfgErr) {
@@ -104,6 +103,9 @@ func assertConfigRejection(t *testing.T, err error, errBuf *bytes.Buffer) {
 	if cfgErr.Subtype != errs.SubtypeInvalidClient {
 		t.Errorf("Subtype = %q, want %q", cfgErr.Subtype, errs.SubtypeInvalidClient)
 	}
+	if cfgErr.Code != wantCode {
+		t.Errorf("Code = %d, want %d", cfgErr.Code, wantCode)
+	}
 	if errBuf.Len() != 0 {
 		t.Errorf("runProbe must not write to stderr, got: %q", errBuf.String())
 	}
@@ -121,13 +123,11 @@ func assertSilent(t *testing.T, err error, errBuf *bytes.Buffer) {
 	}
 }

-// invalid_client (bad / non-existent app_id or wrong secret) → the v3 Token
-// Endpoint returns HTTP 400 with the OAuth2 error → ConfigError/InvalidClient,
-// propagated. The probe endpoint must not be called when TAT fails.
-func TestRunProbe_TATInvalidClient_ReturnsConfigError(t *testing.T) {
+// 10003 (bad / non-existent app_id) → ConfigError/InvalidClient, propagated.
+func TestRunProbe_TATCode10003_ReturnsConfigError(t *testing.T) {
 	rt := &fakeRT{
 		tatHandler: func(req *http.Request) (*http.Response, error) {
-			return jsonResp(400, `{"error":"invalid_client","error_description":"The client secret is invalid.","code":20002}`), nil
+			return jsonResp(200, `{"code":10003,"msg":"invalid param"}`), nil
 		},
 	}
 	f, errBuf := fakeFactory(t, rt)
@@ -137,27 +137,28 @@ func TestRunProbe_TATInvalidClient_ReturnsConfigError(t *testing.T) {
 	if rt.probeCalls != 0 {
 		t.Error("probe endpoint must not be called when TAT fails")
 	}
-	assertConfigRejection(t, err, errBuf)
+	assertConfigRejection(t, err, errBuf, 10003)
 }

-// unauthorized_client is treated as the same credential rejection, propagated.
-func TestRunProbe_TATUnauthorizedClient_ReturnsConfigError(t *testing.T) {
+// 10014 (real app_id + wrong secret) → ConfigError/InvalidClient via codemeta —
+// the most common real-world rejection, propagated.
+func TestRunProbe_TATCode10014_ReturnsConfigError(t *testing.T) {
 	rt := &fakeRT{
 		tatHandler: func(req *http.Request) (*http.Response, error) {
-			return jsonResp(401, `{"error":"unauthorized_client","error_description":"client not authorized"}`), nil
+			return jsonResp(200, `{"code":10014,"msg":"app secret invalid"}`), nil
 		},
 	}
 	f, errBuf := fakeFactory(t, rt)
-	assertConfigRejection(t, runProbe(context.Background(), f, "cli_x", "secret_y", core.BrandFeishu), errBuf)
+	assertConfigRejection(t, runProbe(context.Background(), f, "cli_x", "secret_y", core.BrandFeishu), errBuf, 10014)
 }

-// Any other deterministic client-side OAuth error (e.g. invalid_scope) falls
-// back to *errs.APIError via BuildAPIError — still typed, so the probe surfaces
-// it rather than swallowing — but is not a credential (ConfigError) rejection.
-func TestRunProbe_TATOtherClientError_Propagates(t *testing.T) {
+// Any non-zero body code is a deterministic rejection and propagates (typed).
+// An unrecognized code falls back to *errs.APIError via BuildAPIError — still
+// typed, so the probe still surfaces it rather than swallowing.
+func TestRunProbe_TATUnknownBodyCode_Propagates(t *testing.T) {
 	rt := &fakeRT{
 		tatHandler: func(req *http.Request) (*http.Response, error) {
-			return jsonResp(400, `{"code":20068,"error":"invalid_scope","error_description":"unauthorized scope"}`), nil
+			return jsonResp(200, `{"code":99999,"msg":"future-unknown"}`), nil
 		},
 	}
 	f, errBuf := fakeFactory(t, rt)
--- a/cmd/config/init_test.go
+++ b/cmd/config/init_test.go
@@ -65,8 +65,8 @@ func TestUpdateExistingProfileWithoutSecret_AppIdMismatch_EmitsValidationError(t

 // wrapUpdateExistingProfileErr is the caller-side classifier for the error
 // returned by updateExistingProfileWithoutSecret. It must preserve typed-error
-// exit semantics: a typed ValidationError must keep ExitValidation rather than
-// being downgraded to InternalError.
+// exit semantics (regression: typed ValidationError was being downgraded to
+// InternalError by the legacy *output.ExitError-only passthrough).

 func TestWrapUpdateExistingProfileErr_NilPassesThrough(t *testing.T) {
 	if got := wrapUpdateExistingProfileErr(nil); got != nil {
@@ -90,6 +90,18 @@ func TestWrapUpdateExistingProfileErr_TypedValidationErrorPreserved(t *testing.T
 	}
 }

+func TestWrapUpdateExistingProfileErr_LegacyExitErrorPreserved(t *testing.T) {
+	in := &output.ExitError{Code: 7, Err: errors.New("legacy")}
+	got := wrapUpdateExistingProfileErr(in)
+	var exitErr *output.ExitError
+	if !errors.As(got, &exitErr) {
+		t.Fatalf("expected *output.ExitError to pass through, got %T: %v", got, got)
+	}
+	if exitErr.Code != 7 {
+		t.Errorf("Code = %d, want 7", exitErr.Code)
+	}
+}
+
 func TestWrapUpdateExistingProfileErr_UntypedErrorBecomesInternal(t *testing.T) {
 	in := fmt.Errorf("disk full")
 	got := wrapUpdateExistingProfileErr(in)
--- a/cmd/diagnose_scope_test.go
+++ b/cmd/diagnose_scope_test.go
@@ -47,8 +47,8 @@ func diagAllKnownDomains() []string {
 		seen[p] = true
 	}
 	for _, s := range shortcuts.AllShortcuts() {
-		if s.Service != "" {
-			seen[s.Service] = true
+		if s.GetService() != "" {
+			seen[s.GetService()] = true
 		}
 	}
 	result := make([]string, 0, len(seen))
@@ -94,17 +94,17 @@ func diagBuild(domains []string) diagOutput {
 			}

 			for _, sc := range allSC {
-				if sc.Service != domain || !diagShortcutSupportsIdentity(&sc, identity) {
+				if sc.GetService() != domain || !diagShortcutSupportsIdentity(sc, identity) {
 					continue
 				}
 				for _, scope := range sc.DeclaredScopesForIdentity(identity) {
-					k := methodKey{domain, "shortcut", sc.Command, scope}
+					k := methodKey{domain, "shortcut", sc.GetCommand(), scope}
 					if e, ok := merged[k]; ok {
 						e.Identity = appendUniq(e.Identity, identity)
 					} else {
 						merged[k] = &diagMethodEntry{
 							Domain: domain, Type: "shortcut",
-							Method: sc.Command,
+							Method: sc.GetCommand(),
 							Scope:  scope, Identity: []string{identity},
 						}
 					}
@@ -148,11 +148,12 @@ func diagBuild(domains []string) diagOutput {
 	return diagOutput{Methods: methods, Scopes: scopes}
 }

-func diagShortcutSupportsIdentity(sc *shortcutTypes.Shortcut, identity string) bool {
-	if len(sc.AuthTypes) == 0 {
+func diagShortcutSupportsIdentity(sc shortcutTypes.ShortcutDescriptor, identity string) bool {
+	authTypes := sc.GetAuthTypes()
+	if len(authTypes) == 0 {
 		return identity == "user"
 	}
-	for _, a := range sc.AuthTypes {
+	for _, a := range authTypes {
 		if a == identity {
 			return true
 		}
--- a/cmd/doctor/doctor.go
+++ b/cmd/doctor/doctor.go
@@ -14,7 +14,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/build"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -95,7 +94,7 @@ func doctorRun(opts *DoctorOptions) error {
 		// underlying problem is still visible.
 		msg, hint := err.Error(), ""
 		if errors.Is(err, os.ErrNotExist) {
-			var cfgErr *errs.ConfigError
+			var cfgErr *core.ConfigError
 			if errors.As(core.NotConfiguredError(), &cfgErr) {
 				msg, hint = cfgErr.Message, cfgErr.Hint
 			}
@@ -109,7 +108,7 @@ func doctorRun(opts *DoctorOptions) error {
 	cfg, err := f.Config()
 	if err != nil {
 		hint := ""
-		var cfgErr *errs.ConfigError
+		var cfgErr *core.ConfigError
 		if errors.As(err, &cfgErr) {
 			hint = cfgErr.Hint
 		}
@@ -129,10 +128,7 @@ func doctorRun(opts *DoctorOptions) error {
 	if diagnostics.Bot.Available || diagnostics.User.Available {
 		checks = append(checks, pass("identity_ready", "at least one identity is available"))
 	} else {
-		// No hint: this only summarizes the two checks above, which already carry
-		// the source-appropriate remediation. A command here would be redundant,
-		// or wrong (`auth status` is blocked under an external provider).
-		checks = append(checks, fail("identity_ready", "no usable bot or user identity is available", ""))
+		checks = append(checks, fail("identity_ready", "no usable bot or user identity is available", "run: lark-cli auth status --verify"))
 	}

 	// ── 4 & 5. Endpoint reachability ──
--- a/cmd/doctor/doctor_test.go
+++ b/cmd/doctor/doctor_test.go
@@ -4,19 +4,14 @@
 package doctor

 import (
-	"bytes"
 	"context"
 	"encoding/json"
-	"net/http"
-	"strings"
 	"testing"

 	"github.com/spf13/cobra"

-	extcred "github.com/larksuite/cli/extension/credential"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
-	"github.com/larksuite/cli/internal/credential"
 )

 func TestNewCmdDoctor_FlagParsing(t *testing.T) {
@@ -145,84 +140,14 @@ func TestDoctorRun_SplitsBotAndMissingUserIdentity(t *testing.T) {
 }

 func assertCheck(t *testing.T, checks []checkResult, name, status string) {
-	t.Helper()
-	if got := findCheck(t, checks, name); got.Status != status {
-		t.Fatalf("%s status = %q, want %q", name, got.Status, status)
-	}
-}
-
-func findCheck(t *testing.T, checks []checkResult, name string) checkResult {
 	t.Helper()
 	for _, check := range checks {
 		if check.Name == name {
-			return check
+			if check.Status != status {
+				t.Fatalf("%s status = %q, want %q", name, check.Status, status)
+			}
+			return
 		}
 	}
 	t.Fatalf("check %q not found in %#v", name, checks)
-	return checkResult{}
-}
-
-type fakeExtProvider struct {
-	name    string
-	account *extcred.Account
-}
-
-func (p *fakeExtProvider) Name() string { return p.name }
-func (p *fakeExtProvider) ResolveAccount(context.Context) (*extcred.Account, error) {
-	return p.account, nil
-}
-func (p *fakeExtProvider) ResolveToken(context.Context, extcred.TokenSpec) (*extcred.Token, error) {
-	return nil, nil
-}
-
-// Under an external credential provider with no usable identity, the
-// identity_ready hint must not point at `auth status` (blocked there); the
-// per-identity checks already carry the source-appropriate escalation.
-func TestDoctor_ExternalProvider_IdentityReadyHintNotBlockedCommand(t *testing.T) {
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
-	if err := core.SaveMultiAppConfig(&core.MultiAppConfig{
-		CurrentApp: "default",
-		Apps:       []core.AppConfig{{Name: "default", AppId: "cli_x", AppSecret: core.PlainSecret("secret"), Brand: core.BrandFeishu}},
-	}); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-
-	// Provider serves neither identity: bot unsupported, user supported but not
-	// signed in → both unavailable → identity_ready fails.
-	cfg := &core.CliConfig{AppID: "cli_x", Brand: core.BrandFeishu, SupportedIdentities: uint8(extcred.SupportsUser)}
-	cred := credential.NewCredentialProvider(
-		[]extcred.Provider{&fakeExtProvider{name: "corp-sso", account: &extcred.Account{AppID: "cli_x"}}},
-		nil, nil,
-		func() (*http.Client, error) { return nil, nil },
-	)
-	out := &bytes.Buffer{}
-	f := &cmdutil.Factory{
-		Config:     func() (*core.CliConfig, error) { return cfg, nil },
-		Credential: cred,
-		IOStreams:  &cmdutil.IOStreams{Out: out, ErrOut: &bytes.Buffer{}},
-	}
-
-	if err := doctorRun(&DoctorOptions{Factory: f, Ctx: context.Background(), Offline: true}); err == nil {
-		t.Fatalf("doctorRun() = nil, want failure when no identity is available")
-	}
-	var got struct {
-		Checks []checkResult `json:"checks"`
-	}
-	if err := json.Unmarshal(out.Bytes(), &got); err != nil {
-		t.Fatalf("json.Unmarshal() error = %v\n%s", err, out.String())
-	}
-
-	ready := findCheck(t, got.Checks, "identity_ready")
-	if ready.Status != "fail" {
-		t.Fatalf("identity_ready status = %q, want fail", ready.Status)
-	}
-	// The summary defers to the per-identity checks; it carries no hint of its
-	// own (a command here would be wrong under an external provider).
-	if ready.Hint != "" {
-		t.Fatalf("identity_ready should carry no hint, got %q", ready.Hint)
-	}
-	user := findCheck(t, got.Checks, "user_identity")
-	if !strings.Contains(user.Hint, "external") || strings.Contains(user.Hint, "auth login") {
-		t.Fatalf("user_identity hint not external-appropriate: %q", user.Hint)
-	}
 }
--- a/cmd/error_auth_hint.go
+++ b/cmd/error_auth_hint.go
@@ -11,10 +11,10 @@ import (
 	"github.com/spf13/cobra"

 	"github.com/larksuite/cli/errs"
-	"github.com/larksuite/cli/internal/apicatalog"
 	internalauth "github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
+	"github.com/larksuite/cli/internal/output"
 	"github.com/larksuite/cli/internal/registry"
 	"github.com/larksuite/cli/shortcuts"
 	shortcutcommon "github.com/larksuite/cli/shortcuts/common"
@@ -48,6 +48,32 @@ func applyNeedAuthorizationHint(f *cmdutil.Factory, err error) {
 	authErr.Hint += "\n" + scopeHint
 }

+// enrichMissingScopeError appends a "current command requires scope(s): X"
+// hint to a legacy *output.ExitError when the underlying error carries the
+// need_user_authorization marker AND the current command declares scopes
+// locally.
+//
+// Deprecated: enrichment for the legacy envelope; the typed path is
+// applyNeedAuthorizationHint above.
+func enrichMissingScopeError(f *cmdutil.Factory, exitErr *output.ExitError) {
+	if exitErr == nil || exitErr.Detail == nil {
+		return
+	}
+	if !internalauth.IsNeedUserAuthorizationError(exitErr) {
+		return
+	}
+	scopes := resolveDeclaredScopesForCurrentCommand(f)
+	if len(scopes) == 0 {
+		return
+	}
+	scopeHint := fmt.Sprintf("current command requires scope(s): %s", strings.Join(scopes, ", "))
+	if exitErr.Detail.Hint == "" {
+		exitErr.Detail.Hint = scopeHint
+		return
+	}
+	exitErr.Detail.Hint += "\n" + scopeHint
+}
+
 // resolveDeclaredScopesForCurrentCommand returns the scopes declared by the
 // current command for the resolved identity, checking shortcuts first and then
 // service methods from local registry metadata.
@@ -79,7 +105,7 @@ func resolveDeclaredShortcutScopes(cmd *cobra.Command, identity string) []string

 	service := cmd.Parent().Name()
 	for _, sc := range shortcuts.AllShortcuts() {
-		if sc.Service != service || sc.Command != cmd.Name() || !shortcutSupportsIdentity(sc, identity) {
+		if sc.GetService() != service || sc.GetCommand() != cmd.Name() || !shortcutSupportsIdentity(sc, identity) {
 			continue
 		}
 		scopes := sc.DeclaredScopesForIdentity(identity)
@@ -92,43 +118,44 @@ func resolveDeclaredShortcutScopes(cmd *cobra.Command, identity string) []string
 }

 // resolveDeclaredServiceMethodScopes returns the scopes declared by a
-// service/resource/method command. It reconstructs the catalog path from the
-// command ancestry and resolves it through the same navigation Module the
-// command tree is built from (apicatalog), so it stays correct for nested
-// resources instead of hard-coding a root->service->resource->method depth.
-// Non-method commands (services, resources, shortcuts) resolve to a non-method
-// target and yield no scopes.
+// service/resource/method command from the embedded from_meta registry.
 func resolveDeclaredServiceMethodScopes(cmd *cobra.Command, identity string) []string {
-	if cmd == nil || strings.HasPrefix(cmd.Name(), "+") {
+	// Service-method scope lookup only applies to commands mounted as
+	// root -> service -> resource -> method. Non-resource/method commands
+	// intentionally return no scopes here so auth-hint enrichment does not
+	// change runtime semantics for other command shapes.
+	if cmd == nil || cmd.Parent() == nil || cmd.Parent().Parent() == nil || cmd.Parent().Parent().Parent() == nil {
 		return nil
 	}
-	path := commandCatalogPath(cmd)
-	if len(path) == 0 {
+	if strings.HasPrefix(cmd.Name(), "+") {
 		return nil
 	}
-	target, err := registry.RuntimeCatalog().Resolve(path)
-	if err != nil || target.Kind != apicatalog.TargetMethod {
-		return nil
-	}
-	return registry.DeclaredScopesForMethod(target.Method.Method, identity)
-}

-// commandCatalogPath reconstructs the catalog path [service, resource..., method]
-// from a command's ancestry, excluding the root command. It is the inverse of
-// the service command tree's construction, so any depth (flat or nested)
-// round-trips through apicatalog.Resolve.
-func commandCatalogPath(cmd *cobra.Command) []string {
-	var path []string
-	for c := cmd; c != nil && c.Parent() != nil; c = c.Parent() {
-		path = append([]string{c.Name()}, path...)
+	service := cmd.Parent().Parent().Name()
+	resource := cmd.Parent().Name()
+	method := cmd.Name()
+
+	spec := registry.LoadFromMeta(service)
+	if spec == nil {
+		return nil
 	}
-	return path
+	resources, _ := spec["resources"].(map[string]interface{})
+	resMap, _ := resources[resource].(map[string]interface{})
+	if resMap == nil {
+		return nil
+	}
+	methods, _ := resMap["methods"].(map[string]interface{})
+	methodMap, _ := methods[method].(map[string]interface{})
+	if methodMap == nil {
+		return nil
+	}
+	return registry.DeclaredScopesForMethod(methodMap, identity)
 }

 // shortcutSupportsIdentity reports whether a shortcut supports the requested
 // identity, applying the default user-only behavior when AuthTypes is empty.
-func shortcutSupportsIdentity(sc shortcutcommon.Shortcut, identity string) bool {
-	authTypes := sc.AuthTypes
+func shortcutSupportsIdentity(sc shortcutcommon.ShortcutDescriptor, identity string) bool {
+	authTypes := sc.GetAuthTypes()
 	if len(authTypes) == 0 {
 		authTypes = []string{string(core.AsUser)}
 	}
--- a/cmd/event/appmeta_err.go
+++ b/cmd/event/appmeta_err.go
@@ -8,7 +8,7 @@ import (
 	"regexp"
 )

-// authURLPattern matches the grant-scope URL embedded in 99991672 errors; widen the host alternation when adding brands.
+// authURLPattern matches the grant-scope URL embedded in 99991672 errors; widen when adding brands in consoleScopeGrantURL.
 var authURLPattern = regexp.MustCompile(`https?://open\.(?:feishu\.cn|larksuite\.com)/app/[^/\s"']+/auth\?q=[^\s"'<>]+`)

 // describeAppMetaErr reduces a FetchCurrentPublished error to a one-line stderr summary.
--- a/cmd/event/bus.go
+++ b/cmd/event/bus.go
@@ -12,7 +12,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/event"
@@ -39,8 +38,7 @@ func NewCmdBus(f *cmdutil.Factory) *cobra.Command {

 			logger, err := bus.SetupBusLogger(eventsDir)
 			if err != nil {
-				return errs.NewInternalError(errs.SubtypeFileIO,
-					"set up bus logger: %s", err).WithCause(err)
+				return err
 			}

 			tr := transport.New()
@@ -60,14 +58,7 @@ func NewCmdBus(f *cmdutil.Factory) *cobra.Command {
 				}
 			}()

-			if err := b.Run(ctx); err != nil {
-				if _, ok := errs.ProblemOf(err); ok {
-					return err
-				}
-				return errs.NewInternalError(errs.SubtypeUnknown,
-					"event bus daemon exited: %s", err).WithCause(err)
-			}
-			return nil
+			return b.Run(ctx)
 		},
 	}

--- a/cmd/event/bus_test.go
+++ b/cmd/event/bus_test.go
@@ -1,45 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package event
-
-import (
-	"os"
-	"path/filepath"
-	"testing"
-
-	"github.com/larksuite/cli/errs"
-	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/larksuite/cli/internal/core"
-)
-
-// The hidden `event _bus` daemon command must exit with a typed file_io error
-// when its log directory cannot be created (the error is only visible in the
-// forked process's captured stderr / bus.log).
-func TestBusCommandLoggerSetupFailureIsTypedFileIO(t *testing.T) {
-	dir := t.TempDir()
-	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", dir)
-	// Block the events/ root with a regular file so MkdirAll fails.
-	if err := os.WriteFile(filepath.Join(dir, "events"), []byte("x"), 0600); err != nil {
-		t.Fatal(err)
-	}
-
-	f, _, _, _ := cmdutil.TestFactory(t, &core.CliConfig{
-		AppID: "cli_bus_test", AppSecret: "secret", Brand: core.BrandFeishu,
-	})
-	cmd := NewCmdBus(f)
-	cmd.SetArgs([]string{})
-
-	err := cmd.Execute()
-	if err == nil {
-		t.Fatal("expected logger setup error")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryInternal || p.Subtype != errs.SubtypeFileIO {
-		t.Errorf("problem = %s/%s, want %s/%s", p.Category, p.Subtype,
-			errs.CategoryInternal, errs.SubtypeFileIO)
-	}
-}
--- a/cmd/event/console_url.go
+++ b/cmd/event/console_url.go
@@ -4,117 +4,21 @@
 package event

 import (
-	"bytes"
-	"compress/gzip"
-	"encoding/base64"
-	"encoding/json"
 	"fmt"
+	"strings"

 	"github.com/larksuite/cli/internal/core"
-	eventlib "github.com/larksuite/cli/internal/event"
 )

-// Landing-page contract for the scan-to-enable deep link, verified against the
-// open platform: {open-host}/page/launcher?clientID=<appID>&addons=<encoded>.
-// Note the param is camelCase "clientID" (not snake_case), and the value is the
-// consuming app's own ID. Centralized so it can be corrected in one place.
-const (
-	addonsLandingPath   = "/page/launcher"
-	addonsClientIDParam = "clientID"
-)
-
-// ManifestAddons mirrors the 5 public manifest sections the launcher page accepts.
-// Encoded form: JSON -> gzip -> base64url(no padding).
-type ManifestAddons struct {
-	Scopes    *AddonsScopes    `json:"scopes,omitempty"`
-	Events    *AddonsEvents    `json:"events,omitempty"`
-	Callbacks *AddonsCallbacks `json:"callbacks,omitempty"`
-}
-
-type AddonsScopes struct {
-	Tenant []string `json:"tenant"`
-	User   []string `json:"user"`
-}
-
-type AddonsEvents struct {
-	Items AddonsEventItems `json:"items"`
-}
-
-type AddonsEventItems struct {
-	Tenant []string `json:"tenant"`
-	User   []string `json:"user"`
-}
-
-type AddonsCallbacks struct {
-	Items []string `json:"items"`
-}
-
-// encodeAddons: JSON -> gzip -> base64url(no padding). Matches the front-end decode chain.
-func encodeAddons(a ManifestAddons) (string, error) {
-	raw, err := json.Marshal(a)
-	if err != nil {
-		return "", err
-	}
-	var buf bytes.Buffer
-	gw := gzip.NewWriter(&buf)
-	if _, err := gw.Write(raw); err != nil {
-		return "", err
-	}
-	if err := gw.Close(); err != nil {
-		return "", err
-	}
-	return base64.RawURLEncoding.EncodeToString(buf.Bytes()), nil
-}
-
-// consoleAddonsURL builds the scan-to-enable deep link carrying incremental scopes/events/callbacks.
-func consoleAddonsURL(brand core.LarkBrand, appID string, a ManifestAddons) (string, error) {
-	encoded, err := encodeAddons(a)
-	if err != nil {
-		return "", err
-	}
+// consoleScopeGrantURL builds the developer-console "apply & grant scopes" deep link; scopes are comma-joined without URL encoding.
+func consoleScopeGrantURL(brand core.LarkBrand, appID string, scopes []string) string {
 	host := core.ResolveEndpoints(brand).Open
-	return fmt.Sprintf("%s%s?%s=%s&addons=%s", host, addonsLandingPath, addonsClientIDParam, appID, encoded), nil
+	return fmt.Sprintf("%s/app/%s/auth?q=%s&op_from=openapi&token_type=tenant",
+		host, appID, strings.Join(scopes, ","))
 }

-// consoleLandingURL is the bare landing page (no addons) — fallback when encoding fails.
-func consoleLandingURL(brand core.LarkBrand, appID string) string {
+// consoleEventSubscriptionURL points at the app's event subscription console page.
+func consoleEventSubscriptionURL(brand core.LarkBrand, appID string) string {
 	host := core.ResolveEndpoints(brand).Open
-	return fmt.Sprintf("%s%s?%s=%s", host, addonsLandingPath, addonsClientIDParam, appID)
-}
-
-// addonsHintURL returns the scan URL, degrading to the bare landing page on encode error.
-func addonsHintURL(brand core.LarkBrand, appID string, a ManifestAddons) string {
-	url, err := consoleAddonsURL(brand, appID, a)
-	if err != nil {
-		return consoleLandingURL(brand, appID)
-	}
-	return url
-}
-
-// missingScopeAddons routes missing scopes into the identity-appropriate section.
-// The unused side is an empty (non-nil) slice so JSON encodes [] not null —
-// the addons spec treats a missing tenant/user as an empty array.
-func missingScopeAddons(identity core.Identity, missing []string) ManifestAddons {
-	s := &AddonsScopes{Tenant: []string{}, User: []string{}}
-	if identity.IsBot() {
-		s.Tenant = missing
-	} else {
-		s.User = missing
-	}
-	return ManifestAddons{Scopes: s}
-}
-
-// missingSubscriptionAddons routes missing events/callbacks into the right section.
-// Like missingScopeAddons, unused event sides stay [] (not null) per the addons spec.
-func missingSubscriptionAddons(subType eventlib.SubscriptionType, identity core.Identity, missing []string) ManifestAddons {
-	if subType == eventlib.SubTypeCallback {
-		return ManifestAddons{Callbacks: &AddonsCallbacks{Items: missing}}
-	}
-	ev := &AddonsEvents{Items: AddonsEventItems{Tenant: []string{}, User: []string{}}}
-	if identity.IsBot() {
-		ev.Items.Tenant = missing
-	} else {
-		ev.Items.User = missing
-	}
-	return ManifestAddons{Events: ev}
+	return fmt.Sprintf("%s/app/%s/event", host, appID)
 }
--- a/cmd/event/console_url_test.go
+++ b/cmd/event/console_url_test.go
@@ -4,109 +4,33 @@
 package event

 import (
-	"bytes"
-	"compress/gzip"
-	"encoding/base64"
-	"encoding/json"
-	"io"
-	"strings"
 	"testing"

 	"github.com/larksuite/cli/internal/core"
-	eventlib "github.com/larksuite/cli/internal/event"
 )

-func decodeAddons(t *testing.T, encoded string) ManifestAddons {
-	t.Helper()
-	gz, err := base64.RawURLEncoding.DecodeString(encoded)
-	if err != nil {
-		t.Fatalf("base64url decode: %v", err)
-	}
-	zr, err := gzip.NewReader(bytes.NewReader(gz))
-	if err != nil {
-		t.Fatalf("gzip reader: %v", err)
-	}
-	raw, err := io.ReadAll(zr)
-	if err != nil {
-		t.Fatalf("gunzip: %v", err)
-	}
-	var a ManifestAddons
-	if err := json.Unmarshal(raw, &a); err != nil {
-		t.Fatalf("json: %v", err)
-	}
-	return a
-}
-
-func TestEncodeAddons_RoundTrip(t *testing.T) {
-	in := ManifestAddons{Scopes: &AddonsScopes{Tenant: []string{"im:message"}}}
-	encoded, err := encodeAddons(in)
-	if err != nil {
-		t.Fatalf("encode: %v", err)
-	}
-	for _, r := range encoded {
-		if !(r == '-' || r == '_' || (r >= '0' && r <= '9') || (r >= 'A' && r <= 'Z') || (r >= 'a' && r <= 'z')) {
-			t.Fatalf("encoded contains non-base64url char %q in %q", r, encoded)
-		}
-	}
-	out := decodeAddons(t, encoded)
-	if out.Scopes == nil || len(out.Scopes.Tenant) != 1 || out.Scopes.Tenant[0] != "im:message" {
-		t.Errorf("roundtrip mismatch: %+v", out)
+func TestConsoleScopeGrantURL_Feishu(t *testing.T) {
+	got := consoleScopeGrantURL(core.BrandFeishu, "cli_XXXXXXXXXXXXXXXX", []string{
+		"im:message:readonly",
+		"im:message.group_at_msg",
+	})
+	want := "https://open.feishu.cn/app/cli_XXXXXXXXXXXXXXXX/auth?q=im:message:readonly,im:message.group_at_msg&op_from=openapi&token_type=tenant"
+	if got != want {
+		t.Errorf("url\n got: %s\nwant: %s", got, want)
 	}
 }

-func TestConsoleAddonsURL_FormatAndBrandHost(t *testing.T) {
-	url, err := consoleAddonsURL(core.BrandFeishu, "cli_x", ManifestAddons{Callbacks: &AddonsCallbacks{Items: []string{"card.action.trigger"}}})
-	if err != nil {
-		t.Fatalf("url: %v", err)
-	}
-	host := core.ResolveEndpoints(core.BrandFeishu).Open
-	prefix := host + "/page/launcher?clientID=cli_x&addons="
-	if !strings.HasPrefix(url, prefix) {
-		t.Errorf("url = %q, want prefix %q", url, prefix)
-	}
-	out := decodeAddons(t, strings.TrimPrefix(url, prefix))
-	if out.Callbacks == nil || len(out.Callbacks.Items) != 1 || out.Callbacks.Items[0] != "card.action.trigger" {
-		t.Errorf("decoded callbacks mismatch: %+v", out)
+func TestConsoleScopeGrantURL_LarkBrand(t *testing.T) {
+	got := consoleScopeGrantURL(core.BrandLark, "cli_x", []string{"im:message"})
+	want := "https://open.larksuite.com/app/cli_x/auth?q=im:message&op_from=openapi&token_type=tenant"
+	if got != want {
+		t.Errorf("url\n got: %s\nwant: %s", got, want)
 	}
 }

-func TestMissingScopeAddons_ByIdentity(t *testing.T) {
-	bot := missingScopeAddons(core.AsBot, []string{"im:message"})
-	if bot.Scopes == nil || len(bot.Scopes.Tenant) != 1 || len(bot.Scopes.User) != 0 {
-		t.Errorf("bot scopes = %+v, want tenant-only", bot.Scopes)
-	}
-	user := missingScopeAddons(core.AsUser, []string{"im:message"})
-	if user.Scopes == nil || len(user.Scopes.User) != 1 || len(user.Scopes.Tenant) != 0 {
-		t.Errorf("user scopes = %+v, want user-only", user.Scopes)
-	}
-}
-
-func TestMissingSubscriptionAddons_EventVsCallback(t *testing.T) {
-	ev := missingSubscriptionAddons(eventlib.SubTypeEvent, core.AsBot, []string{"im.message.receive_v1"})
-	if ev.Events == nil || len(ev.Events.Items.Tenant) != 1 {
-		t.Errorf("event addons = %+v, want events.items.tenant", ev.Events)
-	}
-	cb := missingSubscriptionAddons(eventlib.SubTypeCallback, core.AsBot, []string{"card.action.trigger"})
-	if cb.Callbacks == nil || len(cb.Callbacks.Items) != 1 || cb.Events != nil {
-		t.Errorf("callback addons = %+v, want callbacks.items only", cb)
-	}
-}
-
-func TestMissingAddons_EncodeEmptyArraysNotNull(t *testing.T) {
-	// Unused identity sides must encode as [] (not null) so the launcher page's
-	// shape validation treats them as "缺省 -> 空数组" per the addons spec.
-	cases := []ManifestAddons{
-		missingScopeAddons(core.AsBot, []string{"im:message"}),
-		missingScopeAddons(core.AsUser, []string{"im:message"}),
-		missingSubscriptionAddons(eventlib.SubTypeEvent, core.AsBot, []string{"im.message.receive_v1"}),
-	}
-	for i, a := range cases {
-		raw, err := json.Marshal(a)
-		if err != nil {
-			t.Fatalf("case %d marshal: %v", i, err)
-		}
-		if bytes.Contains(raw, []byte("null")) {
-			t.Errorf("case %d encodes a null array, want []: %s", i, raw)
-		}
+func TestConsoleScopeGrantURL_EmptyBrandDefaultsFeishu(t *testing.T) {
+	got := consoleScopeGrantURL("", "cli_x", []string{"im:message"})
+	if got != "https://open.feishu.cn/app/cli_x/auth?q=im:message&op_from=openapi&token_type=tenant" {
+		t.Errorf("unexpected url: %s", got)
 	}
 }
--- a/cmd/event/consume.go
+++ b/cmd/event/consume.go
@@ -16,7 +16,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/appmeta"
 	"github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/cmdutil"
@@ -65,8 +64,8 @@ Use 'event schema <EventKey>' for parameter details.`,
 	cmd.Flags().StringVar(&o.jqExpr, "jq", "", "JQ expression to filter output")
 	cmd.Flags().BoolVar(&o.quiet, "quiet", false, "Suppress informational messages on stderr")
 	cmd.Flags().StringVar(&o.outputDir, "output-dir", "", "Write each event as a file in this directory (relative paths only; absolute paths and ~ are rejected to prevent path traversal)")
-	cmd.Flags().IntVar(&o.maxEvents, "max-events", 0, "Exit after N successful emits (0 = unlimited). Multi-worker EventKeys may emit up to workers-1 past N before all workers stop. Bounded runs ignore stdin EOF.")
-	cmd.Flags().DurationVar(&o.timeout, "timeout", 0, "Exit after DURATION (e.g. 30s, 2m). 0 = no timeout. Timeout is a normal exit (code 0; stderr 'reason: timeout'). Bounded runs ignore stdin EOF.")
+	cmd.Flags().IntVar(&o.maxEvents, "max-events", 0, "Exit after N successful emits (0 = unlimited). Multi-worker EventKeys may emit up to workers-1 past N before all workers stop.")
+	cmd.Flags().DurationVar(&o.timeout, "timeout", 0, "Exit after DURATION (e.g. 30s, 2m). 0 = no timeout. Timeout is a normal exit (code 0; stderr 'reason: timeout').")
 	cmd.Flags().String("as", "auto", "identity type: user | bot | auto (must match EventKey's declared AuthTypes)")
 	_ = cmd.RegisterFlagCompletionFunc("as", func(cmd *cobra.Command, args []string, toComplete string) ([]string, cobra.ShellCompDirective) {
 		return []string{"user", "bot", "auto"}, cobra.ShellCompDirectiveNoFileComp
@@ -102,10 +101,11 @@ func runConsume(cmd *cobra.Command, f *cmdutil.Factory, eventKey string, o consu

 	if o.jqExpr != "" {
 		if err := output.ValidateJqExpression(o.jqExpr); err != nil {
-			return errs.NewValidationError(errs.SubtypeInvalidArgument, "%s", err).
-				WithParam("--jq").
-				WithCause(err).
-				WithHint("see `lark-cli event consume --help` EXAMPLES for common patterns, or `lark-cli event schema %s` for valid field paths", eventKey)
+			return output.ErrWithHint(
+				output.ExitValidation, "validation",
+				err.Error(),
+				fmt.Sprintf("see `lark-cli event consume --help` EXAMPLES for common patterns, or `lark-cli event schema %s` for valid field paths", eventKey),
+			)
 		}
 	}

@@ -146,28 +146,14 @@ func runConsume(cmd *cobra.Command, f *cmdutil.Factory, eventKey string, o consu
 		fmt.Fprintln(preflightErrOut, "[event] skipped console precheck: app has no published version")
 	}

-	// Callback subscriptions live in application/get, not app_versions; fetch the
-	// callback 底账 only for callback-type EventKeys. Weak dependency: on error,
-	// leave subscribedCallbacks nil so the callback precheck skips.
-	var subscribedCallbacks []string
-	if keyDef.SubscriptionType == eventlib.SubTypeCallback {
-		cbs, cbErr := appmeta.FetchSubscribedCallbacks(cmd.Context(), botRuntime, cfg.AppID)
-		if cbErr != nil {
-			fmt.Fprintf(preflightErrOut, "[event] skipped console precheck: %s\n", describeAppMetaErr(cbErr))
-		} else {
-			subscribedCallbacks = cbs
-		}
-	}
-
 	pf := &preflightCtx{
-		factory:             f,
-		appID:               cfg.AppID,
-		brand:               cfg.Brand,
-		eventKey:            eventKey,
-		identity:            identity,
-		keyDef:              keyDef,
-		appVer:              appVer,
-		subscribedCallbacks: subscribedCallbacks,
+		factory:  f,
+		appID:    cfg.AppID,
+		brand:    cfg.Brand,
+		eventKey: eventKey,
+		identity: identity,
+		keyDef:   keyDef,
+		appVer:   appVer,
 	}
 	if err := preflightEventTypes(pf); err != nil {
 		return err
@@ -198,9 +184,8 @@ func runConsume(cmd *cobra.Command, f *cmdutil.Factory, eventKey string, o consu
 		errOut = io.Discard
 	}

-	// Non-TTY unbounded consumers use stdin EOF as shutdown for subprocess callers.
-	// Bounded runs already have --max-events/--timeout as their lifecycle control.
-	if shouldWatchStdinEOF(f.IOStreams.IsTerminal, o.maxEvents, o.timeout) {
+	// Non-TTY only: stdin EOF is shutdown for subprocess callers; in TTY Ctrl-D must not exit.
+	if !f.IOStreams.IsTerminal {
 		watchStdinEOF(os.Stdin, cancel, errOut)
 	}

@@ -243,9 +228,6 @@ type preflightCtx struct {
 	identity core.Identity
 	keyDef   *eventlib.KeyDefinition
 	appVer   *appmeta.AppVersion
-	// subscribedCallbacks is the application/get 底账 for callback-type EventKeys;
-	// nil means "not fetched / unavailable" → callback precheck skips (weak dependency).
-	subscribedCallbacks []string
 }

 // preflightScopes compares required scopes against session-available scopes (user: UAT stored; bot: appVer.TenantScopes).
@@ -278,87 +260,63 @@ func preflightScopes(ctx context.Context, pf *preflightCtx) error {
 	if len(missing) == 0 {
 		return nil
 	}
-	return errs.NewPermissionError(errs.SubtypeMissingScope,
-		"missing required scopes for EventKey %s (as %s): %s",
-		pf.eventKey, pf.identity, strings.Join(missing, ", ")).
-		WithIdentity(string(pf.identity)).
-		WithMissingScopes(missing...).
-		WithHint("%s", scopeRemediationHint(pf.brand, pf.appID, pf.identity, missing))
+	return output.ErrWithHint(
+		output.ExitAuth, "auth",
+		fmt.Sprintf("missing required scopes for EventKey %s (as %s): %s",
+			pf.eventKey, pf.identity, strings.Join(missing, ", ")),
+		scopeRemediationHint(pf.identity, missing, pf.appID, pf.brand),
+	)
 }

 // scopeRemediationHint returns an identity-appropriate fix for missing scopes.
-// Bot: the scan-to-enable link adds the scopes to the app manifest, after which
-// the tenant token carries them. User: the scan link only updates the app
-// manifest — the user's own token still lacks the scopes until it is
-// re-authorized — so direct the user to re-login instead.
-func scopeRemediationHint(brand core.LarkBrand, appID string, identity core.Identity, missing []string) string {
+func scopeRemediationHint(identity core.Identity, missing []string, appID string, brand core.LarkBrand) string {
 	if identity.IsBot() {
-		return fmt.Sprintf("grant these scopes by scanning: %s",
-			addonsHintURL(brand, appID, missingScopeAddons(identity, missing)))
+		return fmt.Sprintf(
+			"grant these scopes and publish a new app version at: %s",
+			consoleScopeGrantURL(brand, appID, missing),
+		)
 	}
 	return fmt.Sprintf(
 		"run `lark-cli auth login --scope \"%s\"` in the background. It blocks and outputs a verification URL — retrieve the URL and open it in a browser to complete login.",
-		strings.Join(missing, " "))
+		strings.Join(missing, " "),
+	)
 }

-// preflightEventTypes verifies every RequiredConsoleEvents entry is subscribed
-// in the app's console 底账 — published app_versions for event subscriptions,
-// application/get subscribed_callbacks for callback subscriptions.
+// preflightEventTypes verifies every RequiredConsoleEvents entry is subscribed in the app's current published version.
 func preflightEventTypes(pf *preflightCtx) error {
-	if len(pf.keyDef.RequiredConsoleEvents) == 0 {
+	if pf.appVer == nil || len(pf.keyDef.RequiredConsoleEvents) == 0 {
 		return nil
 	}
-
-	var subscribed []string
-	noun := "event types"
-	if pf.keyDef.SubscriptionType == eventlib.SubTypeCallback {
-		if pf.subscribedCallbacks == nil {
-			return nil
-		}
-		subscribed = pf.subscribedCallbacks
-		noun = "callbacks"
-	} else {
-		if pf.appVer == nil {
-			return nil
-		}
-		subscribed = pf.appVer.EventTypes
-	}
-
-	have := make(map[string]bool, len(subscribed))
-	for _, t := range subscribed {
-		have[t] = true
+	subscribed := make(map[string]bool, len(pf.appVer.EventTypes))
+	for _, t := range pf.appVer.EventTypes {
+		subscribed[t] = true
 	}
 	var missing []string
 	for _, t := range pf.keyDef.RequiredConsoleEvents {
-		if !have[t] {
+		if !subscribed[t] {
 			missing = append(missing, t)
 		}
 	}
 	if len(missing) == 0 {
 		return nil
 	}
-
-	url := addonsHintURL(pf.brand, pf.appID, missingSubscriptionAddons(pf.keyDef.SubscriptionType, pf.identity, missing))
-	return errs.NewValidationError(errs.SubtypeFailedPrecondition,
-		"EventKey %s requires %s not subscribed in console: %s",
-		pf.keyDef.Key, noun, strings.Join(missing, ", ")).
-		WithHint("subscribe these %s by scanning: %s", noun, url)
+	return output.ErrWithHint(
+		output.ExitValidation, "validation",
+		fmt.Sprintf("EventKey %s requires event types not subscribed in console: %s",
+			pf.keyDef.Key, strings.Join(missing, ", ")),
+		fmt.Sprintf("subscribe these events and publish a new app version at: %s",
+			consoleEventSubscriptionURL(pf.brand, pf.appID)),
+	)
 }

 // sanitizeOutputDir rejects absolute/parent-escaping paths and ~ (SafeOutputPath treats it as a literal dir name).
 func sanitizeOutputDir(dir string) (string, error) {
 	if strings.HasPrefix(dir, "~") {
-		return "", errs.NewValidationError(errs.SubtypeInvalidArgument,
-			"%s; use a relative path like ./output instead", errOutputDirTilde).
-			WithParam("--output-dir").
-			WithCause(errOutputDirTilde)
+		return "", output.ErrValidation("%s; use a relative path like ./output instead", errOutputDirTilde)
 	}
 	safe, err := validate.SafeOutputPath(dir)
 	if err != nil {
-		return "", errs.NewValidationError(errs.SubtypeInvalidArgument,
-			"%s %q: %s", errOutputDirUnsafe, dir, err).
-			WithParam("--output-dir").
-			WithCause(errOutputDirUnsafe)
+		return "", output.ErrValidation("%s %q: %s", errOutputDirUnsafe, dir, err)
 	}
 	return safe, nil
 }
@@ -370,25 +328,22 @@ func resolveTenantToken(ctx context.Context, f *cmdutil.Factory, appID string) (
 	}
 	result, err := f.Credential.ResolveToken(ctx, credential.NewTokenSpec(core.AsBot, appID))
 	if err != nil {
-		if _, ok := errs.ProblemOf(err); ok {
-			return "", err
-		}
-		return "", errs.NewAuthenticationError(errs.SubtypeTokenMissing,
-			"resolve tenant access token: %s", err).WithCause(err)
+		return "", output.ErrAuth("resolve tenant access token: %s", err)
 	}
 	if result == nil || result.Token == "" {
-		return "", errs.NewAuthenticationError(errs.SubtypeTokenMissing,
-			"no tenant access token available for app %s", appID).
-			WithHint("Check that app_secret is configured (lark-cli config show) and try 'lark-cli auth login'.")
+		return "", output.ErrWithHint(
+			output.ExitAuth, "auth",
+			fmt.Sprintf("no tenant access token available for app %s", appID),
+			"Check that app_secret is configured (lark-cli config show) and try 'lark-cli auth login'.",
+		)
 	}
 	return result.Token, nil
 }

-// Sentinels for errors.Is checks; call sites wrap them as typed ValidationError causes.
 var (
-	errInvalidParamFormat = errors.New("invalid --param format")                    //nolint:forbidigo // sentinel, typed at call sites
-	errOutputDirTilde     = errors.New("--output-dir does not support ~ expansion") //nolint:forbidigo // sentinel, typed at call sites
-	errOutputDirUnsafe    = errors.New("unsafe --output-dir")                       //nolint:forbidigo // sentinel, typed at call sites
+	errInvalidParamFormat = errors.New("invalid --param format")
+	errOutputDirTilde     = errors.New("--output-dir does not support ~ expansion")
+	errOutputDirUnsafe    = errors.New("unsafe --output-dir")
 )

 func parseParams(raw []string) (map[string]string, error) {
@@ -396,10 +351,7 @@ func parseParams(raw []string) (map[string]string, error) {
 	for _, kv := range raw {
 		k, v, ok := strings.Cut(kv, "=")
 		if !ok || k == "" {
-			return nil, errs.NewValidationError(errs.SubtypeInvalidArgument,
-				"%s %q: expected key=value", errInvalidParamFormat, kv).
-				WithParam("--param").
-				WithCause(errInvalidParamFormat)
+			return nil, output.ErrValidation("%s %q: expected key=value", errInvalidParamFormat, kv)
 		}
 		m[k] = v
 	}
@@ -418,8 +370,3 @@ func watchStdinEOF(r io.Reader, cancel context.CancelFunc, errOut io.Writer) {
 		cancel()
 	}()
 }
-
-// shouldWatchStdinEOF gates the stdin-EOF shutdown watcher: non-TTY unbounded runs only (<= 0 mirrors downstream's >0-is-bounded semantics, so negative bounds stay unbounded).
-func shouldWatchStdinEOF(isTerminal bool, maxEvents int, timeout time.Duration) bool {
-	return !isTerminal && maxEvents <= 0 && timeout <= 0
-}
--- a/cmd/event/consume_stdin_test.go
+++ b/cmd/event/consume_stdin_test.go
@@ -61,70 +61,3 @@ func TestWatchStdinEOF_DiagnosticMessage(t *testing.T) {
 		t.Fatal("watchStdinEOF did not cancel within 1s of EOF")
 	}
 }
-
-func TestShouldWatchStdinEOF(t *testing.T) {
-	tests := []struct {
-		name       string
-		isTerminal bool
-		maxEvents  int
-		timeout    time.Duration
-		want       bool
-	}{
-		{
-			name:       "terminal",
-			isTerminal: true,
-			want:       false,
-		},
-		{
-			name: "non terminal unbounded",
-			want: true,
-		},
-		{
-			name:      "non terminal negative max events is unbounded",
-			maxEvents: -1,
-			want:      true,
-		},
-		{
-			name:    "non terminal negative timeout is unbounded",
-			timeout: -1 * time.Second,
-			want:    true,
-		},
-		{
-			name:      "non terminal max events bounded",
-			maxEvents: 1,
-			want:      false,
-		},
-		{
-			name:    "non terminal timeout bounded",
-			timeout: 10 * time.Minute,
-			want:    false,
-		},
-		{
-			name:      "non terminal both bounds positive",
-			maxEvents: 1,
-			timeout:   10 * time.Minute,
-			want:      false,
-		},
-		{
-			name:      "non terminal bounded max events with negative timeout",
-			maxEvents: 1,
-			timeout:   -1 * time.Second,
-			want:      false,
-		},
-		{
-			name:      "non terminal bounded timeout with negative max events",
-			maxEvents: -1,
-			timeout:   10 * time.Minute,
-			want:      false,
-		},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			got := shouldWatchStdinEOF(tt.isTerminal, tt.maxEvents, tt.timeout)
-			if got != tt.want {
-				t.Fatalf("shouldWatchStdinEOF() = %v, want %v", got, tt.want)
-			}
-		})
-	}
-}
--- a/cmd/event/consume_test.go
+++ b/cmd/event/consume_test.go
@@ -4,14 +4,9 @@
 package event

 import (
-	"context"
 	"errors"
 	"strings"
 	"testing"
-
-	"github.com/larksuite/cli/errs"
-	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/larksuite/cli/internal/credential"
 )

 func TestParseParams(t *testing.T) {
@@ -78,7 +73,6 @@ func TestParseParams(t *testing.T) {
 				if tc.wantEcho != "" && !strings.Contains(err.Error(), tc.wantEcho) {
 					t.Errorf("err %q should echo %q so user sees the bad input", err.Error(), tc.wantEcho)
 				}
-				assertInvalidArgumentParam(t, err, "--param")
 				return
 			}
 			if err != nil {
@@ -96,77 +90,6 @@ func TestParseParams(t *testing.T) {
 	}
 }

-// emptyTokenResolver resolves to a result that carries no token.
-type emptyTokenResolver struct{}
-
-func (emptyTokenResolver) ResolveToken(_ context.Context, _ credential.TokenSpec) (*credential.TokenResult, error) {
-	return &credential.TokenResult{}, nil
-}
-
-// failingTokenResolver fails outright with an untyped error.
-type failingTokenResolver struct{}
-
-func (failingTokenResolver) ResolveToken(_ context.Context, _ credential.TokenSpec) (*credential.TokenResult, error) {
-	return nil, errors.New("backend unavailable")
-}
-
-func factoryWithResolver(r credential.DefaultTokenResolver) *cmdutil.Factory {
-	return &cmdutil.Factory{Credential: credential.NewCredentialProvider(nil, nil, r, nil)}
-}
-
-func TestResolveTenantToken_EmptyTokenResult(t *testing.T) {
-	_, err := resolveTenantToken(context.Background(), factoryWithResolver(emptyTokenResolver{}), "cli_x")
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryAuthentication || p.Subtype != errs.SubtypeTokenMissing {
-		t.Errorf("problem = %s/%s, want %s/%s", p.Category, p.Subtype,
-			errs.CategoryAuthentication, errs.SubtypeTokenMissing)
-	}
-	var malformed *credential.MalformedTokenResultError
-	if !errors.As(err, &malformed) {
-		t.Error("empty-token failure should preserve the credential-layer cause")
-	}
-}
-
-func TestResolveTenantToken_ResolverFailure(t *testing.T) {
-	_, err := resolveTenantToken(context.Background(), factoryWithResolver(failingTokenResolver{}), "cli_x")
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryAuthentication || p.Subtype != errs.SubtypeTokenMissing {
-		t.Errorf("problem = %s/%s, want %s/%s", p.Category, p.Subtype,
-			errs.CategoryAuthentication, errs.SubtypeTokenMissing)
-	}
-	if errors.Unwrap(err) == nil {
-		t.Error("resolver failure should preserve its cause")
-	}
-}
-
-// assertInvalidArgumentParam verifies err is a typed validation error with
-// subtype invalid_argument naming the given flag in its param field.
-func assertInvalidArgumentParam(t *testing.T, err error, param string) {
-	t.Helper()
-	var ve *errs.ValidationError
-	if !errors.As(err, &ve) {
-		t.Fatalf("expected *errs.ValidationError, got %T: %v", err, err)
-	}
-	if ve.Subtype != errs.SubtypeInvalidArgument {
-		t.Errorf("subtype = %s, want %s", ve.Subtype, errs.SubtypeInvalidArgument)
-	}
-	if ve.Param != param {
-		t.Errorf("param = %q, want %q", ve.Param, param)
-	}
-}
-
 func TestSanitizeOutputDir(t *testing.T) {
 	cases := []struct {
 		name       string
@@ -207,7 +130,6 @@ func TestSanitizeOutputDir(t *testing.T) {
 				if !errors.Is(err, tc.wantSentry) {
 					t.Fatalf("want errors.Is(err, %v), got %q", tc.wantSentry, err.Error())
 				}
-				assertInvalidArgumentParam(t, err, "--output-dir")
 				return
 			}
 			if err != nil {
--- a/cmd/event/format_helpers_test.go
+++ b/cmd/event/format_helpers_test.go
@@ -143,79 +143,6 @@ func TestWriteStatusText_CoversAllStates(t *testing.T) {
 	}
 }

-func TestWriteStatusText_ShowsSubColumn(t *testing.T) {
-	var buf bytes.Buffer
-	writeStatusText(&buf, []appStatus{
-		{
-			AppID:     "cli_RUNNINGXXXXXXXXX",
-			State:     stateRunning,
-			PID:       1234,
-			UptimeSec: 60,
-			Active:    2,
-			Consumers: []protocol.ConsumerInfo{
-				{PID: 1001, EventKey: "mail.x", SubscriptionID: "mail.x:alice", Received: 5, Dropped: 0},
-				{PID: 1002, EventKey: "mail.x", SubscriptionID: "mail.x:bob", Received: 3, Dropped: 0},
-			},
-		},
-	})
-	out := buf.String()
-	if !strings.Contains(out, "SUB") {
-		t.Errorf("missing SUB column header: %s", out)
-	}
-	if !strings.Contains(out, "alice") {
-		t.Errorf("missing alice suffix in SUB column: %s", out)
-	}
-	if !strings.Contains(out, "bob") {
-		t.Errorf("missing bob suffix in SUB column: %s", out)
-	}
-}
-
-func TestWriteStatusText_LegacySubscriptionID_RendersDash(t *testing.T) {
-	var buf bytes.Buffer
-	writeStatusText(&buf, []appStatus{
-		{
-			AppID:     "cli_RUNNINGXXXXXXXXX",
-			State:     stateRunning,
-			PID:       1234,
-			UptimeSec: 60,
-			Active:    1,
-			Consumers: []protocol.ConsumerInfo{
-				{PID: 1001, EventKey: "im.x", SubscriptionID: "", Received: 5},
-			},
-		},
-	})
-	out := buf.String()
-	if !strings.Contains(out, "SUB") {
-		t.Errorf("missing SUB header: %s", out)
-	}
-	if !strings.Contains(out, "-") {
-		t.Errorf("missing dash placeholder for empty SubscriptionID: %s", out)
-	}
-}
-
-func TestWriteStatusText_EventKeyEqualSubscriptionID_RendersDash(t *testing.T) {
-	var buf bytes.Buffer
-	writeStatusText(&buf, []appStatus{
-		{
-			AppID:     "cli_RUNNINGXXXXXXXXX",
-			State:     stateRunning,
-			PID:       1234,
-			UptimeSec: 60,
-			Active:    1,
-			Consumers: []protocol.ConsumerInfo{
-				{PID: 1001, EventKey: "im.x", SubscriptionID: "im.x", Received: 5},
-			},
-		},
-	})
-	out := buf.String()
-	if !strings.Contains(out, "SUB") {
-		t.Errorf("missing SUB header: %s", out)
-	}
-	if !strings.Contains(out, "-") {
-		t.Errorf("missing dash placeholder when SubscriptionID==EventKey: %s", out)
-	}
-}
-
 func TestWriteStatusJSON_OrphanHint(t *testing.T) {
 	var buf bytes.Buffer
 	if err := writeStatusJSON(&buf, []appStatus{
@@ -270,15 +197,15 @@ func TestExitForOrphan(t *testing.T) {
 	if err == nil {
 		t.Fatal("flag on + orphan → expected error, got nil")
 	}
-	var exit *output.BareError
+	var exit *output.ExitError
 	if !errorAs(err, &exit) || exit.Code != output.ExitValidation {
 		t.Errorf("exit code = %v, want ExitValidation", err)
 	}
 }

 func errorAs(err error, target interface{}) bool {
-	if e, ok := err.(*output.BareError); ok {
-		if t, ok := target.(**output.BareError); ok {
+	if e, ok := err.(*output.ExitError); ok {
+		if t, ok := target.(**output.ExitError); ok {
 			*t = e
 			return true
 		}
--- a/cmd/event/list_test.go
+++ b/cmd/event/list_test.go
@@ -10,22 +10,10 @@ import (

 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
-	eventlib "github.com/larksuite/cli/internal/event"

 	_ "github.com/larksuite/cli/events"
 )

-func TestEventLookup_VCMeetingLifecycleKeys(t *testing.T) {
-	for _, key := range []string{
-		"vc.meeting.participant_meeting_started_v1",
-		"vc.meeting.participant_meeting_joined_v1",
-	} {
-		if _, ok := eventlib.Lookup(key); !ok {
-			t.Fatalf("event.Lookup(%q) should succeed", key)
-		}
-	}
-}
-
 func TestRunList_TextOutput(t *testing.T) {
 	f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{AppID: "test"})

@@ -38,9 +26,6 @@ func TestRunList_TextOutput(t *testing.T) {
 		"KEY", "AUTH", "PARAMS", "DESCRIPTION",
 		"im.message.receive_v1",
 		"im.message.message_read_v1",
-		"task.task.update_user_access_v2",
-		"vc.meeting.participant_meeting_started_v1",
-		"vc.meeting.participant_meeting_joined_v1",
 	} {
 		if !strings.Contains(out, want) {
 			t.Errorf("list output missing %q; full output:\n%s", want, out)
@@ -70,31 +55,4 @@ func TestRunList_JSONOutput(t *testing.T) {
 			}
 		}
 	}
-
-	gotKeys := map[string]map[string]interface{}{}
-	for _, row := range rows {
-		if key, ok := row["key"].(string); ok {
-			gotKeys[key] = row
-		}
-	}
-	var foundTask bool
-	for key, row := range gotKeys {
-		if key == "task.task.update_user_access_v2" {
-			foundTask = true
-			if row["single_consumer"] != true {
-				t.Errorf("task row single_consumer = %v, want true", row["single_consumer"])
-			}
-		}
-	}
-	if !foundTask {
-		t.Fatal("event list JSON missing task.task.update_user_access_v2")
-	}
-	for _, want := range []string{
-		"vc.meeting.participant_meeting_started_v1",
-		"vc.meeting.participant_meeting_joined_v1",
-	} {
-		if _, ok := gotKeys[want]; !ok {
-			t.Errorf("JSON list output missing %q", want)
-		}
-	}
 }
--- a/cmd/event/preflight_test.go
+++ b/cmd/event/preflight_test.go
@@ -8,10 +8,10 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/appmeta"
 	"github.com/larksuite/cli/internal/core"
 	eventlib "github.com/larksuite/cli/internal/event"
+	"github.com/larksuite/cli/internal/output"
 )

 func newPreflightCtx(appID string, brand core.LarkBrand, identity core.Identity, keyDef *eventlib.KeyDefinition, appVer *appmeta.AppVersion) *preflightCtx {
@@ -89,17 +89,19 @@ func TestPreflightEventTypes_MissingBlocks(t *testing.T) {
 	if !strings.Contains(err.Error(), "mail.user_mailbox.event.message_read_v1") {
 		t.Errorf("error should name the missing event type, got: %v", err)
 	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
+	var exit *output.ExitError
+	if !errors.As(err, &exit) {
+		t.Fatalf("expected output.ExitError, got %T: %v", err, err)
 	}
-	if p.Category != errs.CategoryValidation || p.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("problem = %s/%s, want %s/%s", p.Category, p.Subtype,
-			errs.CategoryValidation, errs.SubtypeFailedPrecondition)
+	if exit.Code != output.ExitValidation {
+		t.Errorf("ExitCode = %d, want ExitValidation (%d)", exit.Code, output.ExitValidation)
 	}
-	wantURL := "https://open.feishu.cn/page/launcher?clientID=cli_XXXXXXXXXXXXXXXX&addons="
-	if !strings.Contains(p.Hint, wantURL) {
-		t.Errorf("hint missing scan link %q\ngot: %s", wantURL, p.Hint)
+	if exit.Detail == nil {
+		t.Fatal("expected Detail with hint")
+	}
+	wantURL := "https://open.feishu.cn/app/cli_XXXXXXXXXXXXXXXX/event"
+	if !strings.Contains(exit.Detail.Hint, wantURL) {
+		t.Errorf("hint missing subscription URL %q\ngot: %s", wantURL, exit.Detail.Hint)
 	}
 }

@@ -143,22 +145,21 @@ func TestPreflightScopes_Bot_MissingBlocks(t *testing.T) {
 	if !strings.Contains(err.Error(), "im:message.group_at_msg") {
 		t.Errorf("error should name missing scope, got: %v", err)
 	}
-	var permErr *errs.PermissionError
-	if !errors.As(err, &permErr) {
-		t.Fatalf("expected *errs.PermissionError, got %T: %v", err, err)
+	var exit *output.ExitError
+	if !errors.As(err, &exit) {
+		t.Fatalf("expected output.ExitError, got %T: %v", err, err)
 	}
-	if permErr.Category != errs.CategoryAuthorization || permErr.Subtype != errs.SubtypeMissingScope {
-		t.Errorf("problem = %s/%s, want %s/%s", permErr.Category, permErr.Subtype,
-			errs.CategoryAuthorization, errs.SubtypeMissingScope)
+	if exit.Code != output.ExitAuth {
+		t.Errorf("ExitCode = %d, want ExitAuth (%d)", exit.Code, output.ExitAuth)
 	}
-	wantMissing := []string{"im:message.group_at_msg"}
-	if len(permErr.MissingScopes) != 1 || permErr.MissingScopes[0] != wantMissing[0] {
-		t.Errorf("MissingScopes = %v, want %v", permErr.MissingScopes, wantMissing)
+	if exit.Detail == nil {
+		t.Fatal("expected Detail with hint, got nil Detail")
 	}
-	hint := permErr.Hint
+	hint := exit.Detail.Hint
 	wantSubstrings := []string{
-		"grant these scopes by scanning: ",
-		"https://open.feishu.cn/page/launcher?clientID=cli_x&addons=",
+		"https://open.feishu.cn/app/cli_x/auth?q=",
+		"im:message.group_at_msg",
+		"token_type=tenant",
 	}
 	for _, want := range wantSubstrings {
 		if !strings.Contains(hint, want) {
@@ -173,109 +174,3 @@ func TestPreflightScopes_NoRequiredScopes_SkipsCheck(t *testing.T) {
 		t.Fatalf("no required scopes means nothing to verify, got: %v", err)
 	}
 }
-
-func TestPreflightEventTypes_CallbackMissing(t *testing.T) {
-	pf := &preflightCtx{
-		appID:               "cli_x",
-		brand:               core.BrandFeishu,
-		eventKey:            "test.cb",
-		identity:            core.AsBot,
-		subscribedCallbacks: []string{"profile.view.get"},
-		keyDef: &eventlib.KeyDefinition{
-			Key:                   "test.cb",
-			SubscriptionType:      eventlib.SubTypeCallback,
-			RequiredConsoleEvents: []string{"card.action.trigger"},
-		},
-	}
-	err := preflightEventTypes(pf)
-	if err == nil {
-		t.Fatal("expected error for missing callback")
-	}
-	if !strings.Contains(err.Error(), "callbacks not subscribed") {
-		t.Errorf("error = %q, want mention of 'callbacks not subscribed'", err.Error())
-	}
-	if !strings.Contains(err.Error(), "card.action.trigger") {
-		t.Errorf("error should name the missing callback, got: %q", err.Error())
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok || p.Category != errs.CategoryValidation || p.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("problem = %v, want validation/failed_precondition", p)
-	}
-}
-
-func TestPreflightEventTypes_CallbackSkippedWhenNil(t *testing.T) {
-	pf := &preflightCtx{
-		appID:               "cli_x",
-		brand:               core.BrandFeishu,
-		eventKey:            "test.cb",
-		identity:            core.AsBot,
-		subscribedCallbacks: nil, // fetch 失败/拿不到 -> 弱依赖跳过
-		keyDef: &eventlib.KeyDefinition{
-			Key:                   "test.cb",
-			SubscriptionType:      eventlib.SubTypeCallback,
-			RequiredConsoleEvents: []string{"card.action.trigger"},
-		},
-	}
-	if err := preflightEventTypes(pf); err != nil {
-		t.Errorf("expected skip (nil), got %v", err)
-	}
-}
-
-func TestPreflightEventTypes_CallbackEmptyReportsMissing(t *testing.T) {
-	// fetched but zero callbacks subscribed (non-nil empty) is a definitive
-	// console state: a required callback IS missing and must be reported,
-	// not skipped as a weak dependency.
-	pf := &preflightCtx{
-		appID:               "cli_x",
-		brand:               core.BrandFeishu,
-		eventKey:            "test.cb",
-		identity:            core.AsBot,
-		subscribedCallbacks: []string{}, // fetched, none subscribed
-		keyDef: &eventlib.KeyDefinition{
-			Key:                   "test.cb",
-			SubscriptionType:      eventlib.SubTypeCallback,
-			RequiredConsoleEvents: []string{"card.action.trigger"},
-		},
-	}
-	err := preflightEventTypes(pf)
-	if err == nil {
-		t.Fatal("expected error for missing callback when none are subscribed")
-	}
-	if !strings.Contains(err.Error(), "card.action.trigger") {
-		t.Errorf("error should name the missing callback, got: %q", err.Error())
-	}
-}
-
-func TestPreflightEventTypes_CallbackAllSubscribed_Passes(t *testing.T) {
-	pf := &preflightCtx{
-		appID:               "cli_x",
-		brand:               core.BrandFeishu,
-		eventKey:            "test.cb",
-		identity:            core.AsBot,
-		subscribedCallbacks: []string{"card.action.trigger", "profile.view.get"},
-		keyDef: &eventlib.KeyDefinition{
-			Key:                   "test.cb",
-			SubscriptionType:      eventlib.SubTypeCallback,
-			RequiredConsoleEvents: []string{"card.action.trigger"},
-		},
-	}
-	if err := preflightEventTypes(pf); err != nil {
-		t.Errorf("all callbacks subscribed, unexpected error: %v", err)
-	}
-}
-
-func TestScopeRemediationHint_ByIdentity(t *testing.T) {
-	// bot: scan-to-enable link (adds scopes to app manifest)
-	bot := scopeRemediationHint(core.BrandFeishu, "cli_x", core.AsBot, []string{"im:message"})
-	if !strings.Contains(bot, "/page/launcher?clientID=cli_x&addons=") {
-		t.Errorf("bot hint should give the scan link, got: %s", bot)
-	}
-	// user: re-login (scan link cannot grant scopes to the user's own token)
-	user := scopeRemediationHint(core.BrandFeishu, "cli_x", core.AsUser, []string{"im:message"})
-	if !strings.Contains(user, "auth login --scope") {
-		t.Errorf("user hint should direct to auth login, got: %s", user)
-	}
-	if strings.Contains(user, "/page/launcher") {
-		t.Errorf("user hint must NOT use the scan link, got: %s", user)
-	}
-}
--- a/cmd/event/runtime.go
+++ b/cmd/event/runtime.go
@@ -6,8 +6,8 @@ package event
 import (
 	"context"
 	"encoding/json"
+	"fmt"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/client"
 	"github.com/larksuite/cli/internal/core"
 )
@@ -26,11 +26,7 @@ func (r *consumeRuntime) CallAPI(ctx context.Context, method, path string, body
 		As:     r.accessIdentity,
 	})
 	if err != nil {
-		if _, ok := errs.ProblemOf(err); ok {
-			return nil, err
-		}
-		return nil, errs.NewNetworkError(errs.SubtypeNetworkTransport,
-			"api %s %s: %s", method, path, err).WithCause(err)
+		return nil, err
 	}
 	// Non-JSON HTTP errors (gateway text/plain 404 etc.) skip OAPI envelope parsing.
 	ct := resp.Header.Get("Content-Type")
@@ -40,20 +36,11 @@ func (r *consumeRuntime) CallAPI(ctx context.Context, method, path string, body
 		if len(body) > maxBodyEcho {
 			body = body[:maxBodyEcho] + "…(truncated)"
 		}
-		if resp.StatusCode >= 500 {
-			return nil, errs.NewNetworkError(errs.SubtypeNetworkServer,
-				"api %s %s returned %d: %s", method, path, resp.StatusCode, body).WithRetryable()
-		}
-		return nil, errs.NewInternalError(errs.SubtypeInvalidResponse,
-			"api %s %s returned %d: %s", method, path, resp.StatusCode, body)
+		return nil, fmt.Errorf("api %s %s returned %d: %s", method, path, resp.StatusCode, body)
 	}
 	result, err := client.ParseJSONResponse(resp)
 	if err != nil {
-		if _, ok := errs.ProblemOf(err); ok {
-			return nil, err
-		}
-		return nil, errs.NewInternalError(errs.SubtypeInvalidResponse,
-			"api %s %s: %s", method, path, err).WithCause(err)
+		return nil, err
 	}
 	if apiErr := r.client.CheckResponse(result, r.accessIdentity); apiErr != nil {
 		return json.RawMessage(resp.RawBody), apiErr
--- a/cmd/event/runtime_test.go
+++ b/cmd/event/runtime_test.go
@@ -1,147 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package event
-
-import (
-	"context"
-	"errors"
-	"io"
-	"net/http"
-	"strings"
-	"testing"
-
-	lark "github.com/larksuite/oapi-sdk-go/v3"
-	larkcore "github.com/larksuite/oapi-sdk-go/v3/core"
-
-	"github.com/larksuite/cli/errs"
-	"github.com/larksuite/cli/internal/client"
-	"github.com/larksuite/cli/internal/core"
-	"github.com/larksuite/cli/internal/credential"
-)
-
-// staticTokenResolver always returns a fixed token without any HTTP calls.
-type staticTokenResolver struct{}
-
-func (s *staticTokenResolver) ResolveToken(_ context.Context, _ credential.TokenSpec) (*credential.TokenResult, error) {
-	return &credential.TokenResult{Token: "test-token"}, nil
-}
-
-// stubRoundTripper intercepts every outgoing request with a canned response.
-type stubRoundTripper struct {
-	respond func(*http.Request) (*http.Response, error)
-}
-
-func (s stubRoundTripper) RoundTrip(r *http.Request) (*http.Response, error) { return s.respond(r) }
-
-func newTestConsumeRuntime(rt http.RoundTripper) *consumeRuntime {
-	sdk := lark.NewClient("test-app", "test-secret",
-		lark.WithEnableTokenCache(false),
-		lark.WithLogLevel(larkcore.LogLevelError),
-		lark.WithHttpClient(&http.Client{Transport: rt}),
-	)
-	return &consumeRuntime{
-		client: &client.APIClient{
-			SDK:        sdk,
-			ErrOut:     io.Discard,
-			Credential: credential.NewCredentialProvider(nil, nil, &staticTokenResolver{}, nil),
-			Config:     &core.CliConfig{AppID: "test-app", AppSecret: "test-secret", Brand: core.BrandFeishu},
-		},
-		accessIdentity: core.AsBot,
-	}
-}
-
-func stubResponse(status int, contentType, body string) func(*http.Request) (*http.Response, error) {
-	return func(r *http.Request) (*http.Response, error) {
-		return &http.Response{
-			StatusCode: status,
-			Header:     http.Header{"Content-Type": []string{contentType}},
-			Body:       io.NopCloser(strings.NewReader(body)),
-			Request:    r,
-		}, nil
-	}
-}
-
-func requireCallAPIProblem(t *testing.T, err error, category errs.Category, subtype errs.Subtype) {
-	t.Helper()
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != category || p.Subtype != subtype {
-		t.Fatalf("problem = %s/%s, want %s/%s", p.Category, p.Subtype, category, subtype)
-	}
-}
-
-func TestConsumeRuntimeCallAPI_NonJSONHTTPError(t *testing.T) {
-	r := newTestConsumeRuntime(stubRoundTripper{respond: stubResponse(http.StatusNotFound, "text/plain", "gone")})
-	_, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	requireCallAPIProblem(t, err, errs.CategoryInternal, errs.SubtypeInvalidResponse)
-	if !strings.Contains(err.Error(), "returned 404") {
-		t.Errorf("error should echo the HTTP status, got: %v", err)
-	}
-}
-
-func TestConsumeRuntimeCallAPI_NonJSONHTTPErrorTruncatesLongBody(t *testing.T) {
-	long := strings.Repeat("x", 300)
-	r := newTestConsumeRuntime(stubRoundTripper{respond: stubResponse(http.StatusBadGateway, "text/html", long)})
-	_, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	requireCallAPIProblem(t, err, errs.CategoryNetwork, errs.SubtypeNetworkServer)
-	p, _ := errs.ProblemOf(err)
-	if !p.Retryable {
-		t.Fatal("5xx non-JSON response should be marked retryable")
-	}
-	if !strings.Contains(err.Error(), "…(truncated)") {
-		t.Errorf("long body should be truncated in the message, got: %v", err)
-	}
-}
-
-func TestConsumeRuntimeCallAPI_UnparsableJSONBody(t *testing.T) {
-	r := newTestConsumeRuntime(stubRoundTripper{respond: stubResponse(http.StatusOK, "application/json", "{not json")})
-	_, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	requireCallAPIProblem(t, err, errs.CategoryInternal, errs.SubtypeInvalidResponse)
-}
-
-func TestConsumeRuntimeCallAPI_TransportFailure(t *testing.T) {
-	r := newTestConsumeRuntime(stubRoundTripper{respond: func(*http.Request) (*http.Response, error) {
-		return nil, errors.New("connection refused")
-	}})
-	_, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryNetwork {
-		t.Fatalf("category = %s, want %s", p.Category, errs.CategoryNetwork)
-	}
-}
-
-func TestConsumeRuntimeCallAPI_EnvelopeErrorIsTyped(t *testing.T) {
-	r := newTestConsumeRuntime(stubRoundTripper{respond: stubResponse(http.StatusOK, "application/json",
-		`{"code":99991663,"msg":"app not found"}`)})
-	_, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	if _, ok := errs.ProblemOf(err); !ok {
-		t.Fatalf("envelope error should be typed via BuildAPIError, got %T: %v", err, err)
-	}
-}
-
-func TestConsumeRuntimeCallAPI_Success(t *testing.T) {
-	r := newTestConsumeRuntime(stubRoundTripper{respond: stubResponse(http.StatusOK, "application/json",
-		`{"code":0,"data":{"ok":true}}`)})
-	raw, err := r.CallAPI(context.Background(), "GET", "/open-apis/event/v1/connection", nil)
-	if err != nil {
-		t.Fatalf("unexpected error: %v", err)
-	}
-	if !strings.Contains(string(raw), `"code":0`) {
-		t.Errorf("raw body should pass through, got: %s", raw)
-	}
-}
--- a/cmd/event/schema.go
+++ b/cmd/event/schema.go
@@ -11,7 +11,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	eventlib "github.com/larksuite/cli/internal/event"
 	"github.com/larksuite/cli/internal/event/schemas"
@@ -40,14 +39,12 @@ func resolveSchemaJSON(def *eventlib.KeyDefinition) (json.RawMessage, []string,
 	if len(def.Schema.FieldOverrides) > 0 {
 		var parsed map[string]interface{}
 		if err := json.Unmarshal(base, &parsed); err != nil {
-			return nil, nil, errs.NewInternalError(errs.SubtypeUnknown,
-				"parse base schema for field overrides: %s", err).WithCause(err)
+			return nil, nil, err
 		}
 		orphans := schemas.ApplyFieldOverrides(parsed, def.Schema.FieldOverrides)
 		out, err := json.Marshal(parsed)
 		if err != nil {
-			return nil, nil, errs.NewInternalError(errs.SubtypeUnknown,
-				"serialize schema with field overrides: %s", err).WithCause(err)
+			return nil, nil, err
 		}
 		return out, orphans, nil
 	}
@@ -76,7 +73,7 @@ func renderSpec(s *eventlib.SchemaSpec) (json.RawMessage, error) {
 		copy(buf, s.Raw)
 		return buf, nil
 	}
-	return nil, errs.NewInternalError(errs.SubtypeUnknown, "schemaSpec has neither Type nor Raw")
+	return nil, fmt.Errorf("schemaSpec has neither Type nor Raw")
 }

 func NewCmdSchema(f *cmdutil.Factory) *cobra.Command {
@@ -134,16 +131,12 @@ func runSchema(f *cmdutil.Factory, key string, asJSON bool) error {
 	if len(def.Params) > 0 {
 		fmt.Fprintf(out, "\nParameters:\n")
 		w := tabwriter.NewWriter(out, 0, 4, 2, ' ', 0)
-		fmt.Fprintf(w, "  NAME\tTYPE\tREQUIRED\tSUB-KEY\tDEFAULT\tDESCRIPTION\n")
+		fmt.Fprintf(w, "  NAME\tTYPE\tREQUIRED\tDEFAULT\tDESCRIPTION\n")
 		for _, p := range def.Params {
 			required := "no"
 			if p.Required {
 				required = "yes"
 			}
-			subKey := "no"
-			if p.SubscriptionKey {
-				subKey = "yes"
-			}
 			defaultVal := p.Default
 			if defaultVal == "" {
 				defaultVal = "-"
@@ -152,7 +145,7 @@ func runSchema(f *cmdutil.Factory, key string, asJSON bool) error {
 			if desc == "" {
 				desc = "-"
 			}
-			fmt.Fprintf(w, "  %s\t%s\t%s\t%s\t%s\t%s\n", p.Name, p.Type, required, subKey, defaultVal, desc)
+			fmt.Fprintf(w, "  %s\t%s\t%s\t%s\t%s\n", p.Name, p.Type, required, defaultVal, desc)
 		}
 		w.Flush()

@@ -172,7 +165,7 @@ func runSchema(f *cmdutil.Factory, key string, asJSON bool) error {

 	resolved, _, err := resolveSchemaJSON(def)
 	if err != nil {
-		return err
+		return output.Errorf(output.ExitInternal, "internal", "resolve schema: %v", err)
 	}
 	if resolved != nil {
 		fmt.Fprintf(out, "\nOutput Schema:\n")
--- a/cmd/event/schema_test.go
+++ b/cmd/event/schema_test.go
@@ -10,7 +10,6 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	eventlib "github.com/larksuite/cli/internal/event"
@@ -96,146 +95,6 @@ func TestRunSchema_JSONOutput(t *testing.T) {
 	}
 }

-func TestRunSchema_TaskUpdateUserAccessJSON(t *testing.T) {
-	f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{AppID: "test"})
-
-	if err := runSchema(f, "task.task.update_user_access_v2", true); err != nil {
-		t.Fatalf("runSchema json: %v", err)
-	}
-
-	var payload map[string]interface{}
-	if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-		t.Fatalf("output is not valid JSON: %v\n%s", err, stdout.String())
-	}
-	if payload["jq_root_path"] != ".event" {
-		t.Errorf("jq_root_path = %v, want .event", payload["jq_root_path"])
-	}
-	if payload["single_consumer"] != true {
-		t.Errorf("single_consumer = %v, want true", payload["single_consumer"])
-	}
-	resolved := payload["resolved_output_schema"].(map[string]interface{})
-	props := resolved["properties"].(map[string]interface{})
-	eventProps := props["event"].(map[string]interface{})["properties"].(map[string]interface{})
-	if got := eventProps["task_guid"].(map[string]interface{})["format"]; got != "task_guid" {
-		t.Errorf("task_guid format = %v, want task_guid", got)
-	}
-	if _, ok := eventProps["event_types"].(map[string]interface{})["items"].(map[string]interface{})["enum"]; !ok {
-		t.Fatalf("event_types enum missing in schema: %#v", eventProps["event_types"])
-	}
-}
-
-func TestRunSchema_JSONOutput_VCMeetingLifecycleKeys(t *testing.T) {
-	for _, key := range []string{
-		"vc.meeting.participant_meeting_started_v1",
-		"vc.meeting.participant_meeting_joined_v1",
-	} {
-		t.Run(key, func(t *testing.T) {
-			f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{AppID: "test"})
-
-			if err := runSchema(f, key, true); err != nil {
-				t.Fatalf("runSchema json: %v", err)
-			}
-
-			var payload map[string]interface{}
-			if err := json.Unmarshal(stdout.Bytes(), &payload); err != nil {
-				t.Fatalf("output is not valid JSON: %v\n%s", err, stdout.String())
-			}
-			if payload["key"] != key {
-				t.Errorf("key = %v, want %s", payload["key"], key)
-			}
-			resolved, ok := payload["resolved_output_schema"].(map[string]interface{})
-			if !ok {
-				t.Fatalf("resolved_output_schema missing or wrong type: %+v", payload)
-			}
-			properties, ok := resolved["properties"].(map[string]interface{})
-			if !ok {
-				t.Fatalf("resolved_output_schema.properties missing or wrong type: %+v", resolved)
-			}
-			for _, field := range []string{"type", "event_id", "timestamp", "meeting_id", "topic", "meeting_no", "start_time", "calendar_event_id"} {
-				if _, ok := properties[field]; !ok {
-					t.Errorf("resolved output schema missing field %q: %+v", field, properties)
-				}
-			}
-			if _, ok := properties["end_time"]; ok {
-				t.Errorf("resolved output schema should not include end_time for %s: %+v", key, properties)
-			}
-		})
-	}
-}
-
-func TestSchema_RendersSubscriptionKeyMarker(t *testing.T) {
-	const syntheticKey = "test.evt_sub"
-	t.Cleanup(func() { eventlib.UnregisterKeyForTest(syntheticKey) })
-
-	eventlib.RegisterKey(eventlib.KeyDefinition{
-		Key:       syntheticKey,
-		EventType: syntheticKey,
-		Params: []eventlib.ParamDef{
-			{Name: "mailbox", SubscriptionKey: true, Description: "subscription id source"},
-			{Name: "folders", Description: "filter only"},
-		},
-		Schema: eventlib.SchemaDef{Native: &eventlib.SchemaSpec{Type: reflect.TypeOf(struct{ X string }{})}},
-	})
-
-	f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{AppID: "test"})
-	if err := runSchema(f, syntheticKey, false); err != nil {
-		t.Fatalf("runSchema: %v", err)
-	}
-
-	out := stdout.String()
-	if !strings.Contains(out, "SUB-KEY") {
-		t.Errorf("missing SUB-KEY column header in:\n%s", out)
-	}
-
-	// Find the mailbox row and verify "yes" is present
-	var mailboxRow string
-	for _, ln := range strings.Split(out, "\n") {
-		if strings.Contains(ln, "mailbox") && !strings.Contains(ln, "NAME") {
-			mailboxRow = ln
-			break
-		}
-	}
-	if !strings.Contains(mailboxRow, "yes") {
-		t.Errorf("mailbox row missing yes SUB-KEY marker: %q", mailboxRow)
-	}
-
-	// Find the folders row and verify "no" is present
-	var foldersRow string
-	for _, ln := range strings.Split(out, "\n") {
-		if strings.Contains(ln, "folders") && !strings.Contains(ln, "NAME") {
-			foldersRow = ln
-			break
-		}
-	}
-	if !strings.Contains(foldersRow, "no") {
-		t.Errorf("folders row missing no SUB-KEY marker: %q", foldersRow)
-	}
-}
-
-func TestSchema_JSON_IncludesSubscriptionKey(t *testing.T) {
-	const syntheticKey = "test.evt_json"
-	t.Cleanup(func() { eventlib.UnregisterKeyForTest(syntheticKey) })
-
-	eventlib.RegisterKey(eventlib.KeyDefinition{
-		Key:       syntheticKey,
-		EventType: syntheticKey,
-		Params:    []eventlib.ParamDef{{Name: "mailbox", SubscriptionKey: true}},
-		Schema:    eventlib.SchemaDef{Native: &eventlib.SchemaSpec{Type: reflect.TypeOf(struct{ X string }{})}},
-	})
-
-	f, stdout, _, _ := cmdutil.TestFactory(t, &core.CliConfig{AppID: "test"})
-	if err := runSchema(f, syntheticKey, true); err != nil {
-		t.Fatalf("runSchema json: %v", err)
-	}
-
-	if !strings.Contains(stdout.String(), `"subscription_key"`) {
-		t.Errorf("JSON output missing subscription_key field: %s", stdout.String())
-	}
-	if !strings.Contains(stdout.String(), `true`) {
-		t.Errorf("JSON output missing subscription_key: true value: %s", stdout.String())
-	}
-}
-
 func TestResolveSchemaJSON_CustomWithOverlay(t *testing.T) {
 	const syntheticKey = "t.custom.overlay"
 	t.Cleanup(func() { eventlib.UnregisterKeyForTest(syntheticKey) })
@@ -270,38 +129,3 @@ func TestResolveSchemaJSON_CustomWithOverlay(t *testing.T) {
 		t.Errorf("overlay format = %v, want open_id", got)
 	}
 }
-
-func TestRenderSpec_EmptySpecIsTypedInternalError(t *testing.T) {
-	_, err := renderSpec(&eventlib.SchemaSpec{})
-	if err == nil {
-		t.Fatal("expected error for spec with neither Type nor Raw")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryInternal {
-		t.Errorf("category = %s, want %s", p.Category, errs.CategoryInternal)
-	}
-}
-
-func TestResolveSchemaJSON_InvalidBaseWithOverridesIsTypedInternalError(t *testing.T) {
-	def := &eventlib.KeyDefinition{
-		Key: "synthetic.invalid.base",
-		Schema: eventlib.SchemaDef{
-			Custom:         &eventlib.SchemaSpec{Raw: json.RawMessage("{not json")},
-			FieldOverrides: map[string]schemas.FieldMeta{"x": {}},
-		},
-	}
-	_, _, err := resolveSchemaJSON(def)
-	if err == nil {
-		t.Fatal("expected error for unparsable base schema")
-	}
-	p, ok := errs.ProblemOf(err)
-	if !ok {
-		t.Fatalf("expected typed errs error, got %T: %v", err, err)
-	}
-	if p.Category != errs.CategoryInternal {
-		t.Errorf("category = %s, want %s", p.Category, errs.CategoryInternal)
-	}
-}
--- a/cmd/event/status.go
+++ b/cmd/event/status.go
@@ -7,7 +7,6 @@ import (
 	"fmt"
 	"io"
 	"sort"
-	"strings"
 	"sync"
 	"time"

@@ -243,17 +242,12 @@ func writeStatusText(out io.Writer, statuses []appStatus) {
 				s.PID, (time.Duration(s.UptimeSec) * time.Second).String())
 			fmt.Fprintf(out, "  Active consumers: %d\n", s.Active)
 			if len(s.Consumers) > 0 {
-				headers := []string{"CONSUMER", "EVENT KEY", "SUB", "RECEIVED", "DROPPED"}
+				headers := []string{"CONSUMER", "EVENT KEY", "RECEIVED", "DROPPED"}
 				rows := make([][]string, 0, len(s.Consumers))
 				for _, c := range s.Consumers {
-					subDisplay := "-"
-					if c.SubscriptionID != "" && c.SubscriptionID != c.EventKey {
-						subDisplay = strings.TrimPrefix(c.SubscriptionID, c.EventKey+":")
-					}
 					rows = append(rows, []string{
 						fmt.Sprintf("pid=%d", c.PID),
 						c.EventKey,
-						subDisplay,
 						fmt.Sprintf("%d", c.Received),
 						fmt.Sprintf("%d", c.Dropped),
 					})
--- a/cmd/event/status_fail_on_orphan_test.go
+++ b/cmd/event/status_fail_on_orphan_test.go
@@ -19,12 +19,12 @@ func TestExitForOrphan_Orphan(t *testing.T) {
 	if err == nil {
 		t.Fatal("expected error when failOnOrphan=true and orphan present")
 	}
-	var bareErr *output.BareError
-	if !errors.As(err, &bareErr) {
-		t.Fatalf("expected *output.BareError, got %T", err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) {
+		t.Fatalf("expected *output.ExitError, got %T", err)
 	}
-	if bareErr.Code != output.ExitValidation {
-		t.Errorf("Code = %d, want %d", bareErr.Code, output.ExitValidation)
+	if exitErr.Code != output.ExitValidation {
+		t.Errorf("Code = %d, want %d", exitErr.Code, output.ExitValidation)
 	}
 }

--- a/cmd/event/suggestions.go
+++ b/cmd/event/suggestions.go
@@ -8,9 +8,8 @@ import (
 	"sort"
 	"strings"

-	"github.com/larksuite/cli/errs"
 	eventlib "github.com/larksuite/cli/internal/event"
-	"github.com/larksuite/cli/internal/suggest"
+	"github.com/larksuite/cli/internal/output"
 )

 const maxSuggestions = 3
@@ -29,7 +28,7 @@ func suggestEventKeys(input string) []string {
 			hits = append(hits, match{def.Key, 0})
 			continue
 		}
-		if d := suggest.Levenshtein(input, def.Key); d <= threshold {
+		if d := levenshtein(input, def.Key); d <= threshold {
 			hits = append(hits, match{def.Key, d})
 		}
 	}
@@ -64,6 +63,40 @@ func unknownEventKeyErr(key string) error {
 	if guesses := suggestEventKeys(key); len(guesses) > 0 {
 		msg += " — did you mean " + formatSuggestions(guesses) + "?"
 	}
-	return errs.NewValidationError(errs.SubtypeInvalidArgument, "%s", msg).
-		WithHint("Run 'lark-cli event list' to see available keys.")
+	return output.ErrWithHint(
+		output.ExitValidation, "validation",
+		msg,
+		"Run 'lark-cli event list' to see available keys.",
+	)
+}
+
+// levenshtein computes classic edit distance (two-row DP).
+func levenshtein(a, b string) int {
+	if a == b {
+		return 0
+	}
+	ra, rb := []rune(a), []rune(b)
+	if len(ra) == 0 {
+		return len(rb)
+	}
+	if len(rb) == 0 {
+		return len(ra)
+	}
+	prev := make([]int, len(rb)+1)
+	curr := make([]int, len(rb)+1)
+	for j := range prev {
+		prev[j] = j
+	}
+	for i := 1; i <= len(ra); i++ {
+		curr[0] = i
+		for j := 1; j <= len(rb); j++ {
+			cost := 1
+			if ra[i-1] == rb[j-1] {
+				cost = 0
+			}
+			curr[j] = min(prev[j]+1, curr[j-1]+1, prev[j-1]+cost)
+		}
+		prev, curr = curr, prev
+	}
+	return prev[len(rb)]
 }
--- a/cmd/event/suggestions_test.go
+++ b/cmd/event/suggestions_test.go
@@ -10,6 +10,27 @@ import (
 	_ "github.com/larksuite/cli/events"
 )

+func TestLevenshtein(t *testing.T) {
+	cases := []struct {
+		a, b string
+		want int
+	}{
+		{"", "", 0},
+		{"a", "", 1},
+		{"", "abc", 3},
+		{"kitten", "kitten", 0},
+		{"kitten", "sitten", 1},
+		{"kitten", "sitting", 3},
+		{"飞书", "飞书", 0},
+		{"飞书", "飞s", 1},
+	}
+	for _, tc := range cases {
+		if got := levenshtein(tc.a, tc.b); got != tc.want {
+			t.Errorf("levenshtein(%q,%q) = %d, want %d", tc.a, tc.b, got, tc.want)
+		}
+	}
+}
+
 func TestSuggestEventKeys(t *testing.T) {
 	cases := []struct {
 		name              string
--- a/cmd/flag_suggest_test.go
+++ b/cmd/flag_suggest_test.go
@@ -1,104 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd
-
-import (
-	"errors"
-	"strings"
-	"testing"
-
-	"github.com/larksuite/cli/errs"
-	"github.com/larksuite/cli/internal/output"
-	"github.com/spf13/cobra"
-)
-
-func TestUnknownFlagName(t *testing.T) {
-	cases := []struct {
-		in   string
-		name string
-		ok   bool
-	}{
-		{"unknown flag: --query", "query", true},
-		{"unknown flag: --with-styles", "with-styles", true},
-		{"unknown shorthand flag: 'z' in -z", "", false},
-		{"flag needs an argument: --find", "", false},
-		{`invalid argument "x" for "--count"`, "", false},
-	}
-	for _, c := range cases {
-		name, ok := unknownFlagName(errors.New(c.in))
-		if name != c.name || ok != c.ok {
-			t.Errorf("unknownFlagName(%q) = (%q,%v), want (%q,%v)", c.in, name, ok, c.name, c.ok)
-		}
-	}
-}
-
-func TestFlagDidYouMean_UnknownFlagSuggestsAndListsValid(t *testing.T) {
-	c := &cobra.Command{Use: "demo"}
-	c.Flags().String("range", "", "")
-	c.Flags().String("find", "", "")
-	c.Flags().Bool("dry-run", false, "")
-
-	err := flagDidYouMean(c, errors.New("unknown flag: --rang")) // typo of --range
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T", err)
-	}
-	if verr.Subtype != errs.SubtypeInvalidArgument {
-		t.Errorf("subtype = %q, want invalid_argument", verr.Subtype)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	// The offending flag is carried structurally on Params (replaces the
-	// legacy detail map) and named in the message.
-	if len(verr.Params) != 1 || verr.Params[0].Name != "--rang" {
-		t.Errorf("Params = %v, want one entry named --rang", verr.Params)
-	}
-	if len(verr.Params) == 1 && verr.Params[0].Reason == "" {
-		t.Error("Params[0].Reason must explain the rejection")
-	}
-	if !strings.Contains(verr.Message, "--rang") {
-		t.Errorf("message should name the offending flag, got %q", verr.Message)
-	}
-	// The ranked candidate rides on the param as a machine-readable suggestion
-	// so an agent can retry without parsing prose.
-	if len(verr.Params) == 1 {
-		found := false
-		for _, s := range verr.Params[0].Suggestions {
-			if s == "--range" {
-				found = true
-			}
-		}
-		if !found {
-			t.Errorf("Params[0].Suggestions should include --range, got %v", verr.Params[0].Suggestions)
-		}
-	}
-	// The same candidate is also carried in the human-facing hint.
-	if !strings.Contains(verr.Hint, "--range") {
-		t.Errorf("hint should suggest --range, got %q", verr.Hint)
-	}
-}
-
-func TestFlagDidYouMean_OtherErrorStaysGeneric(t *testing.T) {
-	c := &cobra.Command{Use: "demo"}
-	err := flagDidYouMean(c, errors.New("flag needs an argument: --find"))
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T", err)
-	}
-	// Non-unknown-flag errors stay generic: invalid_argument subtype, no
-	// structured param, generic --help hint (no "did you mean" suggestion).
-	if verr.Subtype != errs.SubtypeInvalidArgument {
-		t.Errorf("subtype = %q, want invalid_argument (non-unknown-flag errors stay generic)", verr.Subtype)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	if verr.Param != "" || len(verr.Params) != 0 {
-		t.Errorf("Param=%q Params=%v, want both empty for generic flag error", verr.Param, verr.Params)
-	}
-	if strings.Contains(verr.Hint, "did you mean") {
-		t.Errorf("generic flag error must not produce a did-you-mean hint, got %q", verr.Hint)
-	}
-}
--- a/cmd/notice_test.go
+++ b/cmd/notice_test.go
@@ -1,61 +0,0 @@
-// Copyright (c) 2026 Lark Technologies Pte. Ltd.
-// SPDX-License-Identifier: MIT
-
-package cmd
-
-import (
-	"strings"
-	"testing"
-
-	"github.com/larksuite/cli/internal/deprecation"
-)
-
-// composePendingNotice must surface a deprecated-command alias under the
-// "deprecated_command" key, with the migration target and a skill-update hint,
-// so the JSON "_notice" envelope reaches users who run pre-refactor commands
-// without ever reading --help.
-func TestComposePendingNoticeDeprecatedCommand(t *testing.T) {
-	t.Cleanup(func() { deprecation.SetPending(nil) })
-
-	deprecation.SetPending(&deprecation.Notice{
-		Command:     "+read",
-		Replacement: "+cells-get",
-		Skill:       "lark-sheets",
-	})
-
-	got := composePendingNotice()
-	if got == nil {
-		t.Fatal("composePendingNotice() = nil, want deprecated_command entry")
-	}
-	entry, ok := got["deprecated_command"].(map[string]interface{})
-	if !ok {
-		t.Fatalf("missing deprecated_command key: %#v", got)
-	}
-	if entry["command"] != "+read" {
-		t.Errorf("command = %v, want +read", entry["command"])
-	}
-	if entry["replacement"] != "+cells-get" {
-		t.Errorf("replacement = %v, want +cells-get", entry["replacement"])
-	}
-	if entry["skill"] != "lark-sheets" {
-		t.Errorf("skill = %v, want lark-sheets", entry["skill"])
-	}
-	if msg, _ := entry["message"].(string); !strings.Contains(msg, "update your lark-sheets skill") {
-		t.Errorf("message missing skill-update hint: %q", msg)
-	}
-}
-
-// With nothing pending, the provider returns nil so no "_notice" field is
-// emitted on a clean run.
-func TestComposePendingNoticeEmpty(t *testing.T) {
-	t.Cleanup(func() { deprecation.SetPending(nil) })
-	deprecation.SetPending(nil)
-
-	if got := composePendingNotice(); got != nil {
-		// update/skills pending are process-global; only assert the absence of
-		// our own key to stay robust against unrelated pending state.
-		if _, ok := got["deprecated_command"]; ok {
-			t.Fatalf("deprecated_command present after clear: %#v", got)
-		}
-	}
-}
--- a/cmd/platform_bootstrap_test.go
+++ b/cmd/platform_bootstrap_test.go
@@ -9,12 +9,10 @@ import (
 	"errors"
 	"os"
 	"path/filepath"
-	"strings"
 	"testing"

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/extension/platform"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
@@ -104,7 +102,7 @@ func findLeaf(t *testing.T, parent *cobra.Command, names ...string) *cobra.Comma
 }

 // Happy path: a valid policy.yml denies one specific command. The denied
-// command's RunE returns a typed error envelope; allowed commands are
+// command's RunE returns a typed ExitError envelope; allowed commands are
 // untouched.
 func TestApplyUserPolicyPruning_appliesValidPolicy(t *testing.T) {
 	cfgDir := tmpHome(t)
@@ -129,27 +127,13 @@ max_risk: write
 	if err == nil {
 		t.Fatalf("+delete-doc RunE should return an error")
 	}
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil || exitErr.Detail.Type != "command_denied" {
+		t.Fatalf("expected command_denied ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	// The denial taxonomy (reason_code, layer, rule) is preserved on the
-	// wrapped *platform.CommandDeniedError cause and folded into the hint.
-	var cd *platform.CommandDeniedError
-	if !errors.As(err, &cd) {
-		t.Fatalf("error chain should expose *platform.CommandDeniedError")
-	}
-	if cd.ReasonCode != "command_denylisted" {
-		t.Errorf("CommandDeniedError.ReasonCode = %q, want command_denylisted", cd.ReasonCode)
-	}
-	if !strings.Contains(verr.Hint, "command_denylisted") {
-		t.Errorf("hint should surface reason_code command_denylisted, got %q", verr.Hint)
+	detail, ok := exitErr.Detail.Detail.(map[string]any)
+	if !ok || detail["reason_code"] != "command_denylisted" {
+		t.Errorf("reason_code = %v, want command_denylisted", detail["reason_code"])
 	}

 	// im/+send must be denied (domain not in Allow).
--- a/cmd/platform_guards.go
+++ b/cmd/platform_guards.go
@@ -8,9 +8,9 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/hook"
+	"github.com/larksuite/cli/internal/output"
 	internalplatform "github.com/larksuite/cli/internal/platform"
 )

@@ -34,8 +34,16 @@ import (
 // lands directly on their RunE, which now carries the guard.
 //
 // makeErr is called for every guarded dispatch; it must return a fresh
-// typed error each time.
-func installFatalGuard(rootCmd *cobra.Command, makeErr func() error) {
+// *output.ExitError each time (the envelope writer mutates a few fields
+// as it serialises).
+// Deprecated: installFatalGuard accepts a *output.ExitError-producing lambda,
+// which is part of the legacy error surface that predates the typed error
+// contract introduced by errs/. New code MUST NOT add new callers — the
+// platform-extension fatal-guard plumbing will switch to typed errs.* errors
+// when the platform-extension framework migrates. This wrapper is retained
+// only for the existing in-tree call sites; it will be removed once they
+// have moved to the typed surface.
+func installFatalGuard(rootCmd *cobra.Command, makeErr func() *output.ExitError) {
 	// Two cobra subcommands are injected lazily at Execute() time and
 	// would otherwise slip past walkGuard. We pre-register both so
 	// walkGuard catches them.
@@ -72,65 +80,120 @@ func installFatalGuard(rootCmd *cobra.Command, makeErr func() error) {
 }

 // installPluginInstallErrorGuard surfaces a FailClosed plugin install
-// failure as a typed validation error (failed_precondition) before any
-// command runs.
+// failure as a structured plugin_install envelope before any command
+// runs.
+// Deprecated: installPluginInstallErrorGuard produces a legacy
+// *output.ExitError via its internal makeErr lambda. New code MUST NOT add
+// such producers — plugin install failures should surface as a typed
+// *errs.XxxError once the platform-extension framework migrates. This
+// helper is retained only while existing call sites are migrated; it will
+// be removed once they have moved to the typed surface.
 func installPluginInstallErrorGuard(rootCmd *cobra.Command, installErr error) {
-	makeErr := func() error {
+	makeErr := func() *output.ExitError {
 		var pi *internalplatform.PluginInstallError
 		if errors.As(installErr, &pi) {
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "%s", pi.Error()).
-				WithHint("plugin %q failed to install (reason_code %s); fix or remove the plugin before running commands", pi.PluginName, pi.ReasonCode).
-				WithCause(installErr)
+			return &output.ExitError{
+				Code: output.ExitValidation,
+				Detail: &output.ErrDetail{
+					Type:    "plugin_install",
+					Message: pi.Error(),
+					Detail: map[string]any{
+						"plugin":      pi.PluginName,
+						"reason_code": pi.ReasonCode,
+						"reason":      pi.Reason,
+					},
+				},
+				Err: installErr,
+			}
+		}
+		return &output.ExitError{
+			Code: output.ExitValidation,
+			Detail: &output.ErrDetail{
+				Type:    "plugin_install",
+				Message: installErr.Error(),
+				Detail: map[string]any{
+					"reason_code": internalplatform.ReasonInstallFailed,
+				},
+			},
+			Err: installErr,
 		}
-		return errs.NewValidationError(errs.SubtypeFailedPrecondition, "%s", installErr.Error()).
-			WithHint("a plugin failed to install (reason_code %s); fix or remove the plugin before running commands", internalplatform.ReasonInstallFailed).
-			WithCause(installErr)
 	}
 	installFatalGuard(rootCmd, makeErr)
 }

 // installPluginConflictGuard surfaces a Plugin.Restrict() configuration
 // error (single plugin invalid Rule or multiple plugins each contributing
-// Restrict). The hint separates the two failure modes by reason code:
+// Restrict). The design separates the envelope type:
 //
-//   - "invalid_rule"              - single bad rule
-//   - "multiple_restrict_plugins" - multiple Restrict plugins conflict
+//   - "plugin_install" with reason_code "invalid_rule"           - single bad rule
+//   - "plugin_conflict" with reason_code "multiple_restrict_plugins" - multi
 //
 // Either way the CLI must NOT silently continue with a broken policy.
+// Deprecated: installPluginConflictGuard produces a legacy *output.ExitError
+// via its internal makeErr lambda. New code MUST NOT add such producers —
+// plugin conflict failures should surface as a typed *errs.XxxError once the
+// platform-extension framework migrates. This helper is retained only while
+// existing call sites are migrated; it will be removed once they have moved
+// to the typed surface.
 func installPluginConflictGuard(rootCmd *cobra.Command, err error) {
-	makeErr := func() error {
+	makeErr := func() *output.ExitError {
+		envelopeType := "plugin_install"
 		reasonCode := internalplatform.ReasonInvalidRule
 		if errors.Is(err, cmdpolicy.ErrMultipleRestricts) {
+			envelopeType = "plugin_conflict"
 			reasonCode = internalplatform.ReasonMultipleRestricts
 		}
-		return errs.NewValidationError(errs.SubtypeFailedPrecondition, "%s", err.Error()).
-			WithHint("plugin policy configuration is broken (reason_code %s); fix the plugin's Restrict rule or remove the conflicting plugin", reasonCode).
-			WithCause(err)
+		return &output.ExitError{
+			Code: output.ExitValidation,
+			Detail: &output.ErrDetail{
+				Type:    envelopeType,
+				Message: err.Error(),
+				Detail: map[string]any{
+					"reason_code": reasonCode,
+				},
+			},
+			Err: err,
+		}
 	}
 	installFatalGuard(rootCmd, makeErr)
 }

 // installPluginLifecycleErrorGuard surfaces a Startup lifecycle handler
-// failure as a typed validation error (failed_precondition). The hint's
-// reason code splits returned-error vs panic so consumers (audit /
-// on-call) can tell the two failure modes apart.
+// failure as a plugin_lifecycle envelope. The reason_code splits
+// returned-error vs panic so consumers (audit / on-call) can tell the
+// two failure modes apart.
+// Deprecated: installPluginLifecycleErrorGuard produces a legacy
+// *output.ExitError via its internal makeErr lambda. New code MUST NOT add
+// such producers — plugin lifecycle failures should surface as a typed
+// *errs.XxxError once the platform-extension framework migrates. This
+// helper is retained only while existing call sites are migrated; it will
+// be removed once they have moved to the typed surface.
 func installPluginLifecycleErrorGuard(rootCmd *cobra.Command, err error) {
-	makeErr := func() error {
+	makeErr := func() *output.ExitError {
 		reasonCode := "lifecycle_failed"
-		hookName := ""
+		detail := map[string]any{
+			"reason_code": reasonCode,
+		}
 		var le *hook.LifecycleError
 		if errors.As(err, &le) {
 			if le.Panic {
 				reasonCode = "lifecycle_panic"
 			}
-			hookName = le.HookName
+			detail = map[string]any{
+				"reason_code": reasonCode,
+				"hook_name":   le.HookName,
+				"event":       "startup",
+			}
 		}
-		typed := errs.NewValidationError(errs.SubtypeFailedPrecondition, "%s", err.Error()).
-			WithCause(err)
-		if hookName != "" {
-			return typed.WithHint("plugin startup hook %q failed (reason_code %s); fix or remove the plugin before running commands", hookName, reasonCode)
+		return &output.ExitError{
+			Code: output.ExitValidation,
+			Detail: &output.ErrDetail{
+				Type:    "plugin_lifecycle",
+				Message: err.Error(),
+				Detail:  detail,
+			},
+			Err: err,
 		}
-		return typed.WithHint("a plugin startup hook failed (reason_code %s); fix or remove the plugin before running commands", reasonCode)
 	}
 	installFatalGuard(rootCmd, makeErr)
 }
@@ -156,7 +219,14 @@ func installPluginLifecycleErrorGuard(rootCmd *cobra.Command, err error) {
 //
 // This way the very first non-nil step in cobra's chain is always our
 // guard, regardless of which leaf the user invoked.
-func walkGuard(cmd *cobra.Command, makeErr func() error) {
+// Deprecated: walkGuard accepts a *output.ExitError-producing lambda, part
+// of the legacy error surface that predates the typed error contract
+// introduced by errs/. New code MUST NOT add new callers — the platform-
+// extension guard plumbing will switch to typed errs.* errors when the
+// platform-extension framework migrates. This wrapper is retained only for
+// the existing in-tree call sites; it will be removed once they have moved
+// to the typed surface.
+func walkGuard(cmd *cobra.Command, makeErr func() *output.ExitError) {
 	if cmd == nil {
 		return
 	}
--- a/cmd/platform_guards_test.go
+++ b/cmd/platform_guards_test.go
@@ -6,14 +6,12 @@ package cmd
 import (
 	"context"
 	"errors"
-	"strings"
 	"sync"
 	"testing"
 	"time"

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/extension/platform"
 	"github.com/larksuite/cli/internal/hook"
 	"github.com/larksuite/cli/internal/output"
@@ -34,7 +32,7 @@ func (failClosedAbortingPlugin) Install(platform.Registrar) error {
 }

 // When a FailClosed plugin fails to install, buildInternal must
-// install a PersistentPreRunE that returns a typed *errs.ValidationError.
+// install a PersistentPreRunE that returns a structured *output.ExitError.
 // The user must NEVER see a silent partial-install state.
 //
 // This pins the build.go fix for codex's NEW ISSUE about
@@ -95,31 +93,26 @@ func TestBuildInternal_failClosedAbortsCLI(t *testing.T) {
 	checkGuardError(t, leaf.RunE(leaf, nil))
 }

-// checkGuardError asserts that err is the typed validation error the
-// install guard produces: a failed_precondition *errs.ValidationError
-// (exit 2) whose message + hint preserve the plugin name and the
-// install_failed reason code (the recovery info that lived in the legacy
-// detail map).
+// checkGuardError asserts that err is the structured plugin_install
+// ExitError the guard produces.
 func checkGuardError(t *testing.T, err error) {
 	t.Helper()
 	if err == nil {
 		t.Fatalf("PersistentPreRunE must surface the install error, got nil")
 	}
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "plugin_install" {
+		t.Errorf("envelope type = %q, want plugin_install", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	detail := exitErr.Detail.Detail.(map[string]any)
+	if detail["plugin"] != "policy" {
+		t.Errorf("detail.plugin = %v, want policy", detail["plugin"])
 	}
-	if !strings.Contains(verr.Hint, "policy") {
-		t.Errorf("hint should name the failing plugin %q, got %q", "policy", verr.Hint)
-	}
-	if !strings.Contains(verr.Hint, internalplatform.ReasonInstallFailed) {
-		t.Errorf("hint should surface reason_code %q, got %q", internalplatform.ReasonInstallFailed, verr.Hint)
+	if detail["reason_code"] != internalplatform.ReasonInstallFailed {
+		t.Errorf("detail.reason_code = %v, want install_failed", detail["reason_code"])
 	}
 }

--- a/cmd/plugin_integration_test.go
+++ b/cmd/plugin_integration_test.go
@@ -8,13 +8,11 @@ import (
 	"errors"
 	"os"
 	"path/filepath"
-	"strings"
 	"sync/atomic"
 	"testing"

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/extension/platform"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
@@ -158,23 +156,19 @@ func TestPluginPipeline_wrapAbortReachesEnvelope(t *testing.T) {
 	}

 	err = leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "hook" {
+		t.Errorf("envelope type = %q, want hook", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	detail := exitErr.Detail.Detail.(map[string]any)
+	if detail["reason_code"] != "aborted" {
+		t.Errorf("detail.reason_code = %v, want aborted", detail["reason_code"])
 	}
-	// The namespaced hook name and the abort semantics are preserved in the
-	// message so a caller can identify which plugin hook rejected the call.
-	if !strings.Contains(verr.Message, "policy-plugin.policy") {
-		t.Errorf("message should name the aborting hook policy-plugin.policy, got %q", verr.Message)
-	}
-	if !strings.Contains(verr.Message, "aborted") {
-		t.Errorf("message should describe the abort, got %q", verr.Message)
+	if detail["hook_name"] != "policy-plugin.policy" {
+		t.Errorf("detail.hook_name = %v, want policy-plugin.policy", detail["hook_name"])
 	}

 	// errors.As must still reach the original AbortError so consumers
@@ -415,20 +409,15 @@ func TestPluginConflictGuard_MultipleRestrictAbortsCLI(t *testing.T) {
 		t.Fatalf("no runnable leaf in command tree")
 	}
 	err := leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "plugin_conflict" {
+		t.Errorf("envelope type = %q, want plugin_conflict", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	// reason_code multiple_restrict_plugins is folded into the hint so the
-	// operator can distinguish a multi-Restrict conflict from a bad rule.
-	if !strings.Contains(verr.Hint, "multiple_restrict_plugins") {
-		t.Errorf("hint should surface reason_code multiple_restrict_plugins, got %q", verr.Hint)
+	if rc := exitErr.Detail.Detail.(map[string]any)["reason_code"]; rc != "multiple_restrict_plugins" {
+		t.Errorf("reason_code = %v, want multiple_restrict_plugins", rc)
 	}
 }

@@ -458,20 +447,15 @@ func TestPluginConflictGuard_InvalidRuleAbortsCLI(t *testing.T) {
 		t.Fatalf("no runnable leaf in command tree")
 	}
 	err := leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "plugin_install" {
+		t.Errorf("envelope type = %q, want plugin_install", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	// reason_code invalid_rule is folded into the hint, distinct from the
-	// multiple_restrict_plugins conflict path.
-	if !strings.Contains(verr.Hint, "invalid_rule") {
-		t.Errorf("hint should surface reason_code invalid_rule, got %q", verr.Hint)
+	if rc := exitErr.Detail.Detail.(map[string]any)["reason_code"]; rc != "invalid_rule" {
+		t.Errorf("reason_code = %v, want invalid_rule", rc)
 	}
 }

@@ -500,24 +484,19 @@ func TestPluginLifecycleGuard_StartupErrorAbortsCLI(t *testing.T) {

 	leaf := findRunnableLeaf(root)
 	err := leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "plugin_lifecycle" {
+		t.Errorf("envelope type = %q, want plugin_lifecycle", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	d := exitErr.Detail.Detail.(map[string]any)
+	if d["reason_code"] != "lifecycle_failed" {
+		t.Errorf("reason_code = %v, want lifecycle_failed", d["reason_code"])
 	}
-	// reason_code lifecycle_failed (vs lifecycle_panic) and the failing
-	// hook name are folded into the hint so audit / on-call can tell the
-	// failure mode and which hook failed.
-	if !strings.Contains(verr.Hint, "lifecycle_failed") {
-		t.Errorf("hint should surface reason_code lifecycle_failed, got %q", verr.Hint)
-	}
-	if !strings.Contains(verr.Hint, "lc.start") {
-		t.Errorf("hint should name the failing hook lc.start, got %q", verr.Hint)
+	if d["hook_name"] != "lc.start" {
+		t.Errorf("hook_name = %v, want lc.start", d["hook_name"])
 	}
 }

@@ -541,20 +520,12 @@ func TestPluginLifecycleGuard_StartupPanicAbortsCLI(t *testing.T) {
 	}
 	leaf := findRunnableLeaf(root)
 	err := leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T", err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) {
+		t.Fatalf("expected *output.ExitError, got %T", err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	// A panicking startup hook is distinguished from a returned error by
-	// reason_code lifecycle_panic in the hint.
-	if !strings.Contains(verr.Hint, "lifecycle_panic") {
-		t.Errorf("hint should surface reason_code lifecycle_panic, got %q", verr.Hint)
+	if rc := exitErr.Detail.Detail.(map[string]any)["reason_code"]; rc != "lifecycle_panic" {
+		t.Errorf("reason_code = %v, want lifecycle_panic", rc)
 	}
 }

@@ -608,24 +579,19 @@ func TestWrapperPanic_BecomesHookPanicEnvelope(t *testing.T) {
 	}()

 	err = leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "hook" {
+		t.Errorf("envelope type = %q, want hook", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	d := exitErr.Detail.Detail.(map[string]any)
+	if d["reason_code"] != "panic" {
+		t.Errorf("reason_code = %v, want panic", d["reason_code"])
 	}
-	// The recovered panic surfaces as a structured error naming the
-	// namespaced hook (p.boom) and describing the panic, so the process
-	// never crashes and the caller can attribute the failure.
-	if !strings.Contains(verr.Message, "p.boom") {
-		t.Errorf("message should name the namespaced hook p.boom, got %q", verr.Message)
-	}
-	if !strings.Contains(verr.Message, "panic") {
-		t.Errorf("message should describe the panic, got %q", verr.Message)
+	if d["hook_name"] != "p.boom" {
+		t.Errorf("hook_name = %v, want p.boom (namespaced)", d["hook_name"])
 	}
 }

@@ -687,24 +653,19 @@ func TestWrapperFactoryPanic_BecomesHookPanicEnvelope(t *testing.T) {
 	}()

 	err = leaf.RunE(leaf, nil)
-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("expected *errs.ValidationError, got %T %+v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) || exitErr.Detail == nil {
+		t.Fatalf("expected *output.ExitError, got %T %+v", err, err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if exitErr.Detail.Type != "hook" {
+		t.Errorf("envelope type = %q, want hook", exitErr.Detail.Type)
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	d := exitErr.Detail.Detail.(map[string]any)
+	if d["reason_code"] != "panic" {
+		t.Errorf("reason_code = %v, want panic", d["reason_code"])
 	}
-	// A panic in the wrapper FACTORY (not just the inner handler) is
-	// recovered into the same structured panic error, naming the
-	// namespaced hook fac.bad-factory.
-	if !strings.Contains(verr.Message, "fac.bad-factory") {
-		t.Errorf("message should name the namespaced hook fac.bad-factory, got %q", verr.Message)
-	}
-	if !strings.Contains(verr.Message, "panic") {
-		t.Errorf("message should describe the panic, got %q", verr.Message)
+	if d["hook_name"] != "fac.bad-factory" {
+		t.Errorf("hook_name = %v, want fac.bad-factory (namespaced)", d["hook_name"])
 	}
 }

--- a/cmd/profile/add.go
+++ b/cmd/profile/add.go
@@ -12,7 +12,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/i18n"
@@ -54,9 +53,7 @@ func NewCmdProfileAdd(f *cmdutil.Factory) *cobra.Command {

 func profileAddRun(f *cmdutil.Factory, name, appID string, appSecretStdin bool, brand, lang string, useAfter bool) error {
 	if err := core.ValidateProfileName(name); err != nil {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "%v", err).
-			WithCause(err).
-			WithParam("--name")
+		return output.ErrValidation("%v", err)
 	}

 	langPref, err := cmdutil.ParseLangFlag(lang)
@@ -67,57 +64,46 @@ func profileAddRun(f *cmdutil.Factory, name, appID string, appSecretStdin bool,

 	// Read secret from stdin
 	if !appSecretStdin {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "app secret must be provided via stdin").
-			WithHint("use --app-secret-stdin and pipe the secret").
-			WithParam("--app-secret-stdin")
+		return output.ErrValidation("app secret must be provided via stdin: use --app-secret-stdin and pipe the secret")
 	}
 	scanner := bufio.NewScanner(f.IOStreams.In)
 	if !scanner.Scan() {
 		if err := scanner.Err(); err != nil {
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "failed to read secret from stdin: %v", err).
-				WithCause(err).
-				WithParam("--app-secret-stdin")
+			return output.ErrValidation("failed to read secret from stdin: %v", err)
 		}
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "stdin is empty, expected app secret").
-			WithHint("pipe the app secret to stdin").
-			WithParam("--app-secret-stdin")
+		return output.ErrValidation("stdin is empty, expected app secret")
 	}
 	appSecret := strings.TrimSpace(scanner.Text())
 	if appSecret == "" {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "app secret read from stdin is empty").
-			WithHint("pipe a non-empty app secret to stdin").
-			WithParam("--app-secret-stdin")
+		return output.ErrValidation("app secret read from stdin is empty")
 	}

 	// Load or create config
 	multi, err := core.LoadMultiAppConfig()
 	if err != nil {
 		if !errors.Is(err, os.ErrNotExist) {
-			return errs.NewInternalError(errs.SubtypeFileIO, "failed to load config: %v", err).WithCause(err)
+			return output.Errorf(output.ExitInternal, "internal", "failed to load config: %v", err)
 		}
 		multi = &core.MultiAppConfig{}
 	}

 	// Check name uniqueness
 	if multi.FindApp(name) != nil {
-		return errs.NewValidationError(errs.SubtypeFailedPrecondition, "profile %q already exists", name).
-			WithHint("choose a different name, or remove the existing profile first").
-			WithParam("--name")
+		return output.ErrValidation("profile %q already exists", name)
 	}

 	// Check app-id uniqueness — keychain stores secrets by appId, so
 	// multiple profiles sharing the same appId would collide on credentials.
 	for _, a := range multi.Apps {
 		if a.AppId == appID {
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "app-id %q is already used by profile %q; each profile must have a unique app-id", appID, a.ProfileName()).
-				WithParam("--app-id")
+			return output.ErrValidation("app-id %q is already used by profile %q; each profile must have a unique app-id", appID, a.ProfileName())
 		}
 	}

 	// Store secret securely
 	secret, err := core.ForStorage(appID, core.PlainSecret(appSecret), f.Keychain)
 	if err != nil {
-		return errs.NewInternalError(errs.SubtypeStorage, "%v", err).WithCause(err)
+		return output.Errorf(output.ExitInternal, "internal", "%v", err)
 	}

 	parsedBrand := core.ParseBrand(brand)
@@ -148,7 +134,7 @@ func profileAddRun(f *cmdutil.Factory, name, appID string, appSecretStdin bool,
 	}

 	if err := core.SaveMultiAppConfig(multi); err != nil {
-		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
+		return output.Errorf(output.ExitInternal, "internal", "failed to save config: %v", err)
 	}

 	output.PrintSuccess(f.IOStreams.ErrOut, fmt.Sprintf("Profile %q added (%s, %s)", name, appID, parsedBrand))
--- a/cmd/profile/list.go
+++ b/cmd/profile/list.go
@@ -9,7 +9,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	larkauth "github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -46,7 +45,7 @@ func profileListRun(f *cmdutil.Factory) error {
 			output.PrintJson(f.IOStreams.Out, []profileListItem{})
 			return nil
 		}
-		return errs.NewValidationError(errs.SubtypeFailedPrecondition, "failed to load config: %v", err).WithCause(err)
+		return output.Errorf(output.ExitValidation, "config", "failed to load config: %v", err)
 	}
 	if multi == nil || len(multi.Apps) == 0 {
 		output.PrintJson(f.IOStreams.Out, []profileListItem{})
--- a/cmd/profile/profile_test.go
+++ b/cmd/profile/profile_test.go
@@ -11,7 +11,6 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/i18n"
@@ -51,16 +50,6 @@ func TestProfileAddRun_InvalidExistingConfigReturnsError(t *testing.T) {
 	if !strings.Contains(err.Error(), "failed to load config") {
 		t.Fatalf("error = %v, want failed to load config", err)
 	}
-	var internalErr *errs.InternalError
-	if !errors.As(err, &internalErr) {
-		t.Fatalf("error type = %T, want *errs.InternalError; err=%v", err, err)
-	}
-	if internalErr.Subtype != errs.SubtypeFileIO {
-		t.Fatalf("subtype = %q, want %q", internalErr.Subtype, errs.SubtypeFileIO)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitInternal {
-		t.Fatalf("exit code = %d, want %d (ExitInternal)", code, output.ExitInternal)
-	}
 }

 // TestProfileAddRun_Lang covers the unified --lang contract on profile add:
@@ -106,9 +95,9 @@ func TestProfileAddRun_Lang(t *testing.T) {
 		if err == nil {
 			t.Fatal("expected validation error for --lang ZH, got nil")
 		}
-		var valErr *errs.ValidationError
-		if !errors.As(err, &valErr) || output.ExitCodeOf(err) != output.ExitValidation {
-			t.Fatalf("expected typed validation error with ExitValidation, got %T: %v", err, err)
+		exitErr, ok := err.(*output.ExitError)
+		if !ok || exitErr.Code != output.ExitValidation {
+			t.Fatalf("expected ExitValidation, got %T: %v", err, err)
 		}
 	})
 }
@@ -417,226 +406,17 @@ func TestProfileUseRun_SaveFailureReturnsStructuredError(t *testing.T) {
 func assertInternalExitError(t *testing.T, err error, wantMsg string) {
 	t.Helper()

-	var internalErr *errs.InternalError
-	if !errors.As(err, &internalErr) {
-		t.Fatalf("error type = %T, want *errs.InternalError; err=%v", err, err)
+	var exitErr *output.ExitError
+	if !errors.As(err, &exitErr) {
+		t.Fatalf("error type = %T, want *output.ExitError; err=%v", err, err)
 	}
-	if internalErr.Subtype != errs.SubtypeStorage {
-		t.Fatalf("subtype = %q, want %q", internalErr.Subtype, errs.SubtypeStorage)
+	if exitErr.Code != output.ExitInternal {
+		t.Fatalf("exit code = %d, want %d", exitErr.Code, output.ExitInternal)
 	}
-	if internalErr.Cause == nil {
-		t.Fatalf("cause = nil, want wrapped underlying error")
+	if exitErr.Detail == nil || exitErr.Detail.Type != "internal" {
+		t.Fatalf("detail = %#v, want internal detail", exitErr.Detail)
 	}
-	if !strings.Contains(internalErr.Message, wantMsg) {
-		t.Fatalf("message = %q, want contains %q", internalErr.Message, wantMsg)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitInternal {
-		t.Fatalf("exit code = %d, want %d (ExitInternal)", code, output.ExitInternal)
-	}
-}
-
-// assertValidationError asserts err is a typed *errs.ValidationError with the
-// given subtype, message fragment, and exit code 2.
-func assertValidationError(t *testing.T, err error, wantSubtype errs.Subtype, wantMsg string) *errs.ValidationError {
-	t.Helper()
-
-	if err == nil {
-		t.Fatal("expected error, got nil")
-	}
-	var valErr *errs.ValidationError
-	if !errors.As(err, &valErr) {
-		t.Fatalf("error type = %T, want *errs.ValidationError; err=%v", err, err)
-	}
-	if valErr.Subtype != wantSubtype {
-		t.Fatalf("subtype = %q, want %q", valErr.Subtype, wantSubtype)
-	}
-	if !strings.Contains(valErr.Message, wantMsg) {
-		t.Fatalf("message = %q, want contains %q", valErr.Message, wantMsg)
-	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Fatalf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
-	}
-	return valErr
-}
-
-func saveTwoProfiles(t *testing.T) {
-	t.Helper()
-	multi := &core.MultiAppConfig{
-		CurrentApp: "default",
-		Apps: []core.AppConfig{
-			{Name: "default", AppId: "app-default", AppSecret: core.PlainSecret("secret-default"), Brand: core.BrandFeishu},
-			{Name: "target", AppId: "app-target", AppSecret: core.PlainSecret("secret-target"), Brand: core.BrandLark},
-		},
-	}
-	if err := core.SaveMultiAppConfig(multi); err != nil {
-		t.Fatalf("SaveMultiAppConfig() error = %v", err)
-	}
-}
-
-func TestProfileAddRun_ValidationErrors(t *testing.T) {
-	t.Run("invalid profile name", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		f.IOStreams.In = strings.NewReader("secret\n")
-		err := profileAddRun(f, "bad name!", "app-x", true, "feishu", "", false)
-		valErr := assertValidationError(t, err, errs.SubtypeInvalidArgument, "")
-		if valErr.Param != "--name" {
-			t.Fatalf("param = %q, want %q", valErr.Param, "--name")
-		}
-		if valErr.Cause == nil {
-			t.Fatal("cause = nil, want wrapped validation error")
-		}
-	})
-
-	t.Run("missing app-secret-stdin flag", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileAddRun(f, "p", "app-x", false, "feishu", "", false)
-		valErr := assertValidationError(t, err, errs.SubtypeInvalidArgument, "app secret must be provided via stdin")
-		if valErr.Param != "--app-secret-stdin" {
-			t.Fatalf("param = %q, want %q", valErr.Param, "--app-secret-stdin")
-		}
-		if valErr.Hint == "" {
-			t.Fatal("hint is empty, want actionable hint")
-		}
-	})
-
-	t.Run("empty stdin", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		f.IOStreams.In = strings.NewReader("")
-		err := profileAddRun(f, "p", "app-x", true, "feishu", "", false)
-		valErr := assertValidationError(t, err, errs.SubtypeInvalidArgument, "stdin is empty")
-		if valErr.Param != "--app-secret-stdin" {
-			t.Fatalf("param = %q, want %q", valErr.Param, "--app-secret-stdin")
-		}
-	})
-
-	t.Run("blank secret on stdin", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		f.IOStreams.In = strings.NewReader("   \n")
-		err := profileAddRun(f, "p", "app-x", true, "feishu", "", false)
-		assertValidationError(t, err, errs.SubtypeInvalidArgument, "app secret read from stdin is empty")
-	})
-
-	t.Run("duplicate profile name", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		f.IOStreams.In = strings.NewReader("secret\n")
-		err := profileAddRun(f, "default", "app-new", true, "feishu", "", false)
-		valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, `profile "default" already exists`)
-		if valErr.Param != "--name" {
-			t.Fatalf("param = %q, want %q", valErr.Param, "--name")
-		}
-	})
-
-	t.Run("duplicate app-id", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		f.IOStreams.In = strings.NewReader("secret\n")
-		err := profileAddRun(f, "fresh", "app-default", true, "feishu", "", false)
-		valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, "already used by profile")
-		if valErr.Param != "--app-id" {
-			t.Fatalf("param = %q, want %q", valErr.Param, "--app-id")
-		}
-	})
-}
-
-func TestProfileUseRun_ValidationErrors(t *testing.T) {
-	t.Run("no previous profile for toggle", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileUseRun(f, "-")
-		valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, "no previous profile to switch back to")
-		if valErr.Hint == "" {
-			t.Fatal("hint is empty, want actionable hint")
-		}
-	})
-
-	t.Run("profile not found", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileUseRun(f, "ghost")
-		assertValidationError(t, err, errs.SubtypeInvalidArgument, `profile "ghost" not found`)
-	})
-}
-
-func TestProfileRenameRun_ValidationErrors(t *testing.T) {
-	t.Run("invalid new name", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileRenameRun(f, "default", "bad name!")
-		valErr := assertValidationError(t, err, errs.SubtypeInvalidArgument, "")
-		if valErr.Cause == nil {
-			t.Fatal("cause = nil, want wrapped validation error")
-		}
-	})
-
-	t.Run("old profile not found", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileRenameRun(f, "ghost", "fresh")
-		assertValidationError(t, err, errs.SubtypeInvalidArgument, `profile "ghost" not found`)
-	})
-
-	t.Run("new name already exists", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileRenameRun(f, "default", "target")
-		valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, `profile "target" already exists`)
-		if valErr.Hint == "" {
-			t.Fatal("hint is empty, want actionable hint")
-		}
-	})
-}
-
-func TestProfileRemoveRun_ValidationErrors(t *testing.T) {
-	t.Run("profile not found", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		saveTwoProfiles(t)
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileRemoveRun(f, "ghost")
-		assertValidationError(t, err, errs.SubtypeInvalidArgument, `profile "ghost" not found`)
-	})
-
-	t.Run("cannot remove the only profile", func(t *testing.T) {
-		setupProfileConfigDir(t)
-		multi := &core.MultiAppConfig{
-			CurrentApp: "solo",
-			Apps: []core.AppConfig{
-				{Name: "solo", AppId: "app-solo", AppSecret: core.PlainSecret("secret-solo"), Brand: core.BrandFeishu},
-			},
-		}
-		if err := core.SaveMultiAppConfig(multi); err != nil {
-			t.Fatalf("SaveMultiAppConfig() error = %v", err)
-		}
-		f, _, _, _ := cmdutil.TestFactory(t, nil)
-		err := profileRemoveRun(f, "solo")
-		valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, "cannot remove the only profile")
-		if valErr.Hint == "" {
-			t.Fatal("hint is empty, want actionable hint")
-		}
-	})
-}
-
-func TestProfileListRun_InvalidConfigReturnsValidationError(t *testing.T) {
-	dir := setupProfileConfigDir(t)
-	if err := os.WriteFile(filepath.Join(dir, "config.json"), []byte("{invalid json"), 0600); err != nil {
-		t.Fatalf("WriteFile() error = %v", err)
-	}
-
-	f, _, _, _ := cmdutil.TestFactory(t, nil)
-	err := profileListRun(f)
-	valErr := assertValidationError(t, err, errs.SubtypeFailedPrecondition, "failed to load config")
-	if valErr.Cause == nil {
-		t.Fatal("cause = nil, want wrapped load error")
+	if !strings.Contains(exitErr.Detail.Message, wantMsg) {
+		t.Fatalf("message = %q, want contains %q", exitErr.Detail.Message, wantMsg)
 	}
 }
--- a/cmd/profile/remove.go
+++ b/cmd/profile/remove.go
@@ -9,7 +9,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	larkauth "github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
@@ -41,12 +40,11 @@ func profileRemoveRun(f *cmdutil.Factory, name string) error {

 	idx := multi.FindAppIndex(name)
 	if idx < 0 {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "profile %q not found, available profiles: %s", name, strings.Join(multi.ProfileNames(), ", "))
+		return output.ErrValidation("profile %q not found, available profiles: %s", name, strings.Join(multi.ProfileNames(), ", "))
 	}

 	if len(multi.Apps) == 1 {
-		return errs.NewValidationError(errs.SubtypeFailedPrecondition, "cannot remove the only profile").
-			WithHint("add another profile first: lark-cli profile add")
+		return output.ErrValidation("cannot remove the only profile")
 	}

 	app := &multi.Apps[idx]
@@ -67,7 +65,7 @@ func profileRemoveRun(f *cmdutil.Factory, name string) error {
 	}

 	if err := core.SaveMultiAppConfig(multi); err != nil {
-		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
+		return output.Errorf(output.ExitInternal, "internal", "failed to save config: %v", err)
 	}

 	// Best-effort credential cleanup after config commit
--- a/cmd/profile/rename.go
+++ b/cmd/profile/rename.go
@@ -9,7 +9,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/output"
@@ -31,7 +30,7 @@ func NewCmdProfileRename(f *cmdutil.Factory) *cobra.Command {

 func profileRenameRun(f *cmdutil.Factory, oldName, newName string) error {
 	if err := core.ValidateProfileName(newName); err != nil {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "%v", err).WithCause(err)
+		return output.ErrValidation("%v", err)
 	}

 	multi, err := core.LoadOrNotConfigured()
@@ -41,7 +40,7 @@ func profileRenameRun(f *cmdutil.Factory, oldName, newName string) error {

 	idx := multi.FindAppIndex(oldName)
 	if idx < 0 {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "profile %q not found, available profiles: %s", oldName, strings.Join(multi.ProfileNames(), ", "))
+		return output.ErrValidation("profile %q not found, available profiles: %s", oldName, strings.Join(multi.ProfileNames(), ", "))
 	}

 	// Check new name uniqueness across other profiles, allowing renames to this
@@ -51,8 +50,7 @@ func profileRenameRun(f *cmdutil.Factory, oldName, newName string) error {
 			continue
 		}
 		if multi.Apps[i].Name == newName || multi.Apps[i].AppId == newName {
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "profile %q already exists", newName).
-				WithHint("choose a different name")
+			return output.ErrValidation("profile %q already exists", newName)
 		}
 	}

@@ -68,7 +66,7 @@ func profileRenameRun(f *cmdutil.Factory, oldName, newName string) error {
 	}

 	if err := core.SaveMultiAppConfig(multi); err != nil {
-		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
+		return output.Errorf(output.ExitInternal, "internal", "failed to save config: %v", err)
 	}

 	output.PrintSuccess(f.IOStreams.ErrOut, fmt.Sprintf("Profile renamed: %q -> %q", oldProfileName, newName))
--- a/cmd/profile/use.go
+++ b/cmd/profile/use.go
@@ -9,7 +9,6 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
 	"github.com/larksuite/cli/internal/output"
@@ -41,15 +40,14 @@ func profileUseRun(f *cmdutil.Factory, name string) error {
 	// Handle "-" for toggle-back
 	if name == "-" {
 		if multi.PreviousApp == "" {
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "no previous profile to switch back to").
-				WithHint("switch to a profile by name first: lark-cli profile use <name>")
+			return output.ErrValidation("no previous profile to switch back to")
 		}
 		name = multi.PreviousApp
 	}

 	app := multi.FindApp(name)
 	if app == nil {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "profile %q not found, available profiles: %s", name, strings.Join(multi.ProfileNames(), ", "))
+		return output.ErrValidation("profile %q not found, available profiles: %s", name, strings.Join(multi.ProfileNames(), ", "))
 	}

 	targetName := app.ProfileName()
@@ -68,7 +66,7 @@ func profileUseRun(f *cmdutil.Factory, name string) error {
 	multi.CurrentApp = targetName

 	if err := core.SaveMultiAppConfig(multi); err != nil {
-		return errs.NewInternalError(errs.SubtypeStorage, "failed to save config: %v", err).WithCause(err)
+		return output.Errorf(output.ExitInternal, "internal", "failed to save config: %v", err)
 	}

 	output.PrintSuccess(f.IOStreams.ErrOut, fmt.Sprintf("Switched to profile %q (%s, %s)", targetName, app.AppId, app.Brand))
--- a/cmd/prune.go
+++ b/cmd/prune.go
@@ -9,10 +9,10 @@ import (

 	"github.com/spf13/cobra"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
+	"github.com/larksuite/cli/internal/output"
 )

 // pruneForStrictMode removes commands incompatible with the active strict mode.
@@ -65,10 +65,10 @@ func strictModeStubFrom(child *cobra.Command, mode core.StrictMode) *cobra.Comma
 	//     pick auth's instead of our denial. A leaf-level no-op makes
 	//     cobra stop here and proceed to the wrapped RunE.
 	//
-	// strict-mode keeps its short Message + independent Hint and wraps
-	// the CommandDeniedError as the Cause by hand; BuildDenialError
-	// would override Message with the CommandDeniedError.Error() long
-	// form.
+	// strict-mode keeps its short Message + independent Hint and
+	// composes the shared detail.* / wrapped-CommandDeniedError shape
+	// by hand; BuildDenialError would override Message with the
+	// CommandDeniedError.Error() long form.
 	stubMessage := fmt.Sprintf(
 		"strict mode is %q, only %s-identity commands are available",
 		mode, mode.ForcedIdentity())
@@ -105,9 +105,20 @@ func strictModeStubFrom(child *cobra.Command, mode core.StrictMode) *cobra.Comma
 		},
 		RunE: func(c *cobra.Command, _ []string) error {
 			cd := cmdpolicy.CommandDeniedFromDenial(cmdpolicy.CanonicalPath(c), denial)
-			return errs.NewValidationError(errs.SubtypeFailedPrecondition, "%s", stubMessage).
-				WithHint("denied by %s policy (reason_code %s); %s", cd.Layer, cd.ReasonCode, stubHint).
-				WithCause(cd)
+			// Legacy *output.ExitError producer: this literal predates the
+			// typed error contract introduced by errs/. New denial sites MUST
+			// NOT construct *output.ExitError directly — they should return a
+			// typed *errs.XxxError once the cmdpolicy framework migrates.
+			return &output.ExitError{
+				Code: output.ExitValidation,
+				Detail: &output.ErrDetail{
+					Type:    "command_denied",
+					Message: stubMessage,
+					Hint:    stubHint,
+					Detail:  cmdpolicy.DenialDetailMap(cd),
+				},
+				Err: cd,
+			}
 		},
 	}
 }
--- a/cmd/prune_test.go
+++ b/cmd/prune_test.go
@@ -8,7 +8,6 @@ import (
 	"strings"
 	"testing"

-	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/extension/platform"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
@@ -248,12 +247,9 @@ func TestStrictModeStub_BypassesArgsValidator(t *testing.T) {
 	}
 }

-// Pins the strict-mode typed envelope: a failed_precondition
-// *errs.ValidationError (exit 2) carrying the short historical Message,
-// a Hint that still surfaces the policy layer + reason code (the
-// safety-critical recovery info that lived in the legacy detail map),
-// and the wrapped *platform.CommandDeniedError so external agents can
-// still inspect the structured denial taxonomy via errors.As.
+// Pins the strict-mode envelope shape: structured detail.* / wrapped
+// CommandDeniedError for external agents, AND the historical short
+// Message + independent Hint for existing consumers.
 func TestStrictModeStub_StructuredEnvelope(t *testing.T) {
 	root := newTestTree()
 	pruneForStrictMode(root, core.StrictModeBot)
@@ -266,33 +262,30 @@ func TestStrictModeStub_StructuredEnvelope(t *testing.T) {
 		t.Fatalf("strict-mode stub RunE should return error")
 	}

-	var verr *errs.ValidationError
-	if !errors.As(err, &verr) {
-		t.Fatalf("err is not *errs.ValidationError: %T", err)
+	var ee *output.ExitError
+	if !errors.As(err, &ee) {
+		t.Fatalf("err is not *output.ExitError: %T", err)
 	}
-	if verr.Subtype != errs.SubtypeFailedPrecondition {
-		t.Errorf("subtype = %q, want failed_precondition", verr.Subtype)
+	if ee.Detail == nil {
+		t.Fatalf("ExitError.Detail is nil; envelope writer cannot emit JSON")
 	}
-	if code := output.ExitCodeOf(err); code != output.ExitValidation {
-		t.Errorf("exit code = %d, want %d (ExitValidation)", code, output.ExitValidation)
+	if ee.Detail.Type != "command_denied" {
+		t.Errorf("Detail.Type = %q, want command_denied", ee.Detail.Type)
 	}
-	// Short historical Message is preserved verbatim.
-	if verr.Message != `strict mode is "bot", only bot-identity commands are available` {
-		t.Errorf("Message = %q, want short historical form", verr.Message)
+	dm, ok := ee.Detail.Detail.(map[string]any)
+	if !ok {
+		t.Fatalf("Detail.Detail = %T, want map[string]any", ee.Detail.Detail)
 	}
-	// The denial layer + reason code remain user-readable in the hint, and
-	// the historical switch-policy guidance is still appended.
-	if !strings.Contains(verr.Hint, cmdpolicy.LayerStrictMode) {
-		t.Errorf("Hint = %q, want substring %q (policy layer)", verr.Hint, cmdpolicy.LayerStrictMode)
+	if got, _ := dm["layer"].(string); got != cmdpolicy.LayerStrictMode {
+		t.Errorf("Detail.Detail[layer] = %q, want %q", got, cmdpolicy.LayerStrictMode)
 	}
-	if !strings.Contains(verr.Hint, "identity_not_supported") {
-		t.Errorf("Hint = %q, want substring identity_not_supported (reason code)", verr.Hint)
+	if got, _ := dm["reason_code"].(string); got != "identity_not_supported" {
+		t.Errorf("Detail.Detail[reason_code] = %q, want identity_not_supported", got)
 	}
-	if !strings.Contains(verr.Hint, "if the user explicitly wants to switch policy") {
-		t.Errorf("Hint = %q, want historical switch-policy guidance", verr.Hint)
+	if got, _ := dm["policy_source"].(string); got != "strict-mode" {
+		t.Errorf("Detail.Detail[policy_source] = %q, want strict-mode", got)
 	}

-	// The structured denial taxonomy survives on the wrapped cause.
 	var cd *platform.CommandDeniedError
 	if !errors.As(err, &cd) {
 		t.Fatalf("err does not unwrap to *platform.CommandDeniedError")
@@ -303,12 +296,15 @@ func TestStrictModeStub_StructuredEnvelope(t *testing.T) {
 	if cd.ReasonCode != "identity_not_supported" {
 		t.Errorf("CommandDeniedError.ReasonCode = %q, want identity_not_supported", cd.ReasonCode)
 	}
-	if cd.PolicySource != "strict-mode" {
-		t.Errorf("CommandDeniedError.PolicySource = %q, want strict-mode", cd.PolicySource)
-	}
 	if !strings.Contains(cd.Reason, `strict mode is "bot"`) {
 		t.Errorf("CommandDeniedError.Reason = %q, want substring 'strict mode is \"bot\"'", cd.Reason)
 	}
+	if ee.Detail.Message != `strict mode is "bot", only bot-identity commands are available` {
+		t.Errorf("Detail.Message = %q, want short historical form", ee.Detail.Message)
+	}
+	if !strings.HasPrefix(ee.Detail.Hint, "if the user explicitly wants to switch policy") {
+		t.Errorf("Detail.Hint = %q, want historical hint", ee.Detail.Hint)
+	}
 }

 // strictModeStubFrom must write the denial annotations so the hook
--- a/cmd/root.go
+++ b/cmd/root.go
@@ -11,90 +11,65 @@ import (
 	"sort"
 	"strings"

-	"github.com/larksuite/cli/cmd/service"
 	"github.com/larksuite/cli/errs"
 	"github.com/larksuite/cli/extension/platform"
+	internalauth "github.com/larksuite/cli/internal/auth"
 	"github.com/larksuite/cli/internal/build"
-	"github.com/larksuite/cli/internal/cmdmeta"
 	"github.com/larksuite/cli/internal/cmdpolicy"
 	"github.com/larksuite/cli/internal/cmdutil"
-	"github.com/larksuite/cli/internal/deprecation"
+	"github.com/larksuite/cli/internal/core"
+	"github.com/larksuite/cli/internal/errclass"
+	"github.com/larksuite/cli/internal/errcompat"
 	"github.com/larksuite/cli/internal/hook"
 	"github.com/larksuite/cli/internal/output"
+	"github.com/larksuite/cli/internal/registry"
 	"github.com/larksuite/cli/internal/skillscheck"
-	"github.com/larksuite/cli/internal/suggest"
 	"github.com/larksuite/cli/internal/update"
 	"github.com/spf13/cobra"
-	"github.com/spf13/pflag"
 )

 const rootLong = `lark-cli — Lark/Feishu CLI tool.

-AGENT QUICKSTART (driving this as an agent? start here):
-    Browse commands:  lark-cli <domain> --help            # +shortcuts (preferred) and raw API resources
-    Inspect a call:   lark-cli schema <service>.<resource>.<method>   # params, types, scopes, examples
-    Prefer a +shortcut over the raw API resource when one matches the task.
-    Risk: each command's --help shows read | write | high-risk-write;
-          high-risk-write needs --yes, only after the user confirms.
-    On any API call: --jq <expr> filters JSON output, --dry-run previews the request (runs nothing).
+USAGE:
+    lark-cli <command> [subcommand] [method] [options]
+    lark-cli api <method> <path> [--params <json>] [--data <json>]
+    lark-cli schema <service.resource.method> [--format pretty]

-EXAMPLES (one per command style, in order of preference):
-    lark-cli calendar +agenda                                       # +shortcut — a high-level task, prefer these
-    lark-cli mail user_mailbox.messages list --user-mailbox-id me   # typed command for one API method
-    lark-cli schema mail.user_mailbox.messages.list                 # inspect a method's params before calling
-    lark-cli api GET /open-apis/calendar/v4/calendars               # raw escape hatch — any endpoint by HTTP path`
+EXAMPLES:
+    # View upcoming events
+    lark-cli calendar +agenda

-// rootUsageTemplate is cobra's default usage template with two root-only
-// additions gated on {{if not .HasParent}}: a curated multi-form Usage synopsis
-// (replacing cobra's generic "[flags] / [command]") and a human skills-setup
-// footer. Subcommands render the stock template unchanged. The rest is verbatim
-// cobra so the command groups and flags are untouched.
-const rootUsageTemplate = `{{if .HasParent}}Usage:{{if .Runnable}}
-  {{.UseLine}}{{end}}{{if .HasAvailableSubCommands}}
-  {{.CommandPath}} [command]{{end}}{{else}}Usage:
-  lark-cli <command> [subcommand] [method] [flags]
-  lark-cli api <method> <path> [--params <json>] [--data <json>]
-  lark-cli schema <service.resource.method>{{end}}{{if gt (len .Aliases) 0}}
+    # List calendar events
+    lark-cli calendar events instance_view --params '{"calendar_id":"primary","start_time":"1700000000","end_time":"1700086400"}'

-Aliases:
-  {{.NameAndAliases}}{{end}}{{if .HasExample}}
+    # Search users
+    lark-cli contact +search-user --query "John"

-Examples:
-{{.Example}}{{end}}{{if .HasAvailableSubCommands}}{{$cmds := .Commands}}{{if eq (len .Groups) 0}}
+    # Generic API call
+    lark-cli api GET /open-apis/calendar/v4/calendars

-Available Commands:{{range $cmds}}{{if (or .IsAvailableCommand (eq .Name "help"))}}
-  {{rpad .Name .NamePadding }} {{.Short}}{{end}}{{end}}{{else}}{{range $group := .Groups}}
+AI AGENT SKILLS:
+    lark-cli pairs with AI agent skills (Claude Code, etc.) that
Author	SHA1	Message	Date
shanglei	5cd34b1f70	Merge branch 'main' into feat/shortcut-protocol	2026-06-03 13:50:45 +08:00
shanglei	9a53a1f2b8	fix(shortcuts): typed []string named-element binding + nil-Execute mount parity Two parity edge cases from a follow-up audit: - []<named string> element types (type ID string; field []ID) passed parse validation but panicked at bind via reflect.Convert([]string -> []ID). Build the slice element-by-element with SetString (stringsToSlice) so plain []string, named slice types and named element types all bind correctly. - nil Execute: mountTyped now skips mounting (matching legacy Shortcut.MountWithContext) instead of mounting a command that errors only at invocation, keeping the command tree identical after migration. Adds binder_namedslice_nilexec_test.go (4 cases).	2026-05-29 15:30:20 +08:00
shanglei	eb6f5aa60a	feat(shortcuts): typed []string flags, per-flag hidden, @file help hint Close the last legacy/typed capability gaps so migration never downgrades: - []string fields: registerLeaf/bindLeaf gain a Slice case. Default maps to cobra StringSlice (comma-separated, repeatable); `split:"none"` opts into StringArray (repeatable, no comma split). Non-[]string slices and split on non-slice fields error at Mount time. bucketLeafValue handles slices in OneOf/group too. - per-flag hidden: `hidden:"true"` tag → fieldSpec.Hidden → MarkHidden; typed help skips hidden flags (parity with legacy Flag.Hidden). - @file/stdin discoverability: typed help now appends a (supports @file / - for stdin) hint for input-tagged flags, matching legacy. Adds binder_slice_hidden_test.go (11 cases).	2026-05-29 15:15:09 +08:00
shanglei	c4eb18cecc	feat(shortcuts): typed @file/stdin input + enum completion/help parity Bring TypedShortcut[T] to parity with legacy common.Shortcut on two flag capabilities that were silently missing on the typed path: - @file / stdin: declare `input:"file,stdin"` on an Args field; the binder resolves @path (file) and - (stdin) before binding, recursing into OneOf buckets and groups. resolveInputForFlag is extracted from the legacy resolveInputFlags so both paths behave identically. - enum: registerLeaf registers shell completion and typed help renders the candidate list, matching legacy. enum/input tags on non-string fields now error at Mount time instead of being silently skipped at runtime. Legacy behavior unchanged. Adds binder_input_enum_test.go (11 cases).	2026-05-29 14:55:40 +08:00
shanglei	a510e07dfc	refactor(shortcuts): drop unused argstype.UserOpenIDList UserOpenIDList ([]string) had zero production callers and cannot be bound as a typed flag field: the binder only handles string/bool/int scalars, so a []string field panics at bind time (reflect.Value.Convert: string cannot convert to []string). Remove the type, its ParseUserOpenIDList helper and the test instead of special-casing slice binding for a type nobody uses.	2026-05-29 11:51:16 +08:00
shanglei	f83c79825d	refactor(shortcuts): retire Maybe[T] — top-level T pointer covers tri-state Maybe[T] solved a real problem (distinguishing "user did not provide this flag" from "user provided the type's zero value, e.g. --notify=false or --limit 0"), but it added a framework-specific wrapper that every migrator had to learn. Two independent dogfood migrations both flagged it as the guide's biggest cognitive bump. This change retires Maybe[T] and reuses the existing OneOf pointer convention at the top level: T now means "optional with tri-state semantics — nil iff the user did not set the flag". The rule is the same whether the T sits inside a OneOf bucket (Chat ChatID — variant not selected) or at the top level of an Args struct (Notify bool — flag not given), so users learn one concept instead of two. Code: - binder.go: bindLeaf now gates pointer allocation behind cmd.Flags().Changed(name); previously a T top-level pointer was always allocated, defeating the "nil = not given" semantic. fieldSpec loses the IsMaybe field; parseFieldSpec drops Maybe-shape detection; registerFlags / bindFlags drop their IsMaybe branches; the bindMaybe helper is deleted. - protocol.go: Maybe[T] type removed. - protocol_test.go: TestMaybe_UnsetVsZero removed. - binder_coverage_test.go: bindMaybe tests replaced with two TestBindLeaf_Ptr* tests covering the new "*T top-level pointer" contract — nil when absent, non-nil with the user's value (including the tri-state case --notify=false) when explicitly set. Verified: go build ./..., go test ./shortcuts/common/..., gofmt clean, and `go run golang.org/x/tools/cmd/deadcode@v0.31.0 -test ./...` vs origin/main shows zero new dead code.	2026-05-28 19:29:34 +08:00
shanglei	9adc79d0c1	fix(shortcuts): typed help shows REQUIRED + defaults; restore TypedShortcut.Mount Three fixes surfaced by a dogfood migration of an existing legacy shortcut to the TypedShortcut framework. 1. Required flags rendered under OPTIONAL The typed help renderer had no notion of required vs optional and lumped every top-level leaf into a single OPTIONAL section. Add renderRequiredSection that lists fields whose `required` tag is set under a REQUIRED: header; the updated renderOptionalSection skips those so each leaf appears in exactly one section. 2. Default values silently dropped Fields tagged `default:"x"` showed no `(default "x")` suffix, unlike cobra's legacy default help. Add a formatLeafLine helper that appends `(default "x")` whenever fieldSpec.DefaultValue is non-empty; the REQUIRED, OPTIONAL, and CHOOSE ONE bucket renderers all reuse it for consistent information density. 3. TypedShortcut.Mount was missing Legacy common.Shortcut exposes BOTH .Mount(parent, factory) and .MountWithContext. The framework dropped .Mount in `8d8acb82` to satisfy deadcode, but a dogfood migration revealed the asymmetry forces every migrating shortcut's tests to also switch to MountWithContext. Restore the 3-line Mount convenience method (delegates to MountWithContext with a background context) and add a unit test exercising it directly — the test doubles as API documentation and keeps deadcode happy. Tests added: - TestTypedShortcut_Mount: verifies the legacy-shaped Mount(parent, factory) call still wires the subcommand - TestTypedHelp_RequiredSectionAndDefaults: asserts REQUIRED comes before OPTIONAL, --limit lands in REQUIRED not OPTIONAL, `(default "20")` appears for --page-size, and a plain --verbose carries no default suffix Verified locally: go build ./..., go test ./shortcuts/common/..., gofmt -l, and `go run golang.org/x/tools/cmd/deadcode@v0.31.0 -test ./...` vs origin/main — all pass with zero new dead code.	2026-05-28 17:24:58 +08:00
shanglei	6b4bc0cc64	style(shortcuts): gofmt binder_test.go CI fast-gate's `gofmt -l .` flagged an alignment issue in contentBucket: gofmt aligns struct field tags to the longest named tag-bearing field in the group; my hand-written extra spaces in `Text *string \`flag:"ct"\`` got normalised to a single space. Behaviour-neutral; tests still pass.	2026-05-28 16:51:17 +08:00
shanglei	b5cd535285	refactor(shortcuts): drop oneof_trigger tag, infer variant attempt from any inner flag The previous checkOneOf required group / bucket variants inside a OneOf to mark exactly one inner field with `oneof_trigger:"true"`, so the framework could identify "the flag that signals the variant was selected". This forces the Args writer to think about an implementation detail and produces a misleading shortcut_oneof_missing when a user supplies only a companion field (e.g. --video-cover without --video) — the user actually attempted that variant, but the framework couldn't see it. Replace the explicit trigger with implicit detection: any inner flag of a group / bucket variant that is Changed counts as attempting that variant. The follow-up checkGroup catches the partial-fill case and surfaces a shortcut_group_incomplete pointing at the missing flag, which is strictly more useful than the prior oneof_missing. Code changes: - binder.go: drop fieldSpec.OneOfTrig, drop parseFieldSpec's oneof_trigger tag handling, rewrite checkOneOf with the inferred-attempt logic, delete the isTrigger helper (now inlined and simplified). - typed_help.go: renderFlagsInBucket flattens all inner flags to the parent's indent (no more "trigger vs companion" visual distinction, matching the framework's new semantics). - binder_test.go: add three behavioural tests covering the new contract: (1) companion alone surfaces group_incomplete (not oneof_missing), (2) full group passes, (3) simple variant + group companion both attempted yields oneof_multiple. Verified: go build ./..., go test ./shortcuts/common/..., `go run golang.org/x/tools/cmd/deadcode@v0.31.0 -test ./...` against HEAD vs origin/main yields zero new dead code. Net: -1 tag, -1 field on fieldSpec, -1 helper function, simpler writer ergonomics, more accurate error messages.	2026-05-28 16:42:40 +08:00
shanglei	098659cc18	feat(shortcuts): bind top-level group sub-struct fields in TypedShortcut bindFlags and bindBuckets only populated top-level leaves and OneOf buckets; a top-level group sub-struct (a regular nested struct without OneOf() marker) had its inner flags registered and group-completeness checked, but its field values were never written back into the Args struct — Execute would read empty values from args.Group.X. Add bindGroups as the top-level counterpart to bindBuckets for IsGroup fields, and invoke it from mountTyped's Validate closure right after bindBuckets. Behavior mirrors bindBuckets / bindBucketInner: - Value-type top-level group: always populated; inner fields receive cobra flag values (including defaults). - Pointer-type top-level group: allocated iff at least one inner flag was Changed, so a nil group still signals "user did not engage this group" while a non-nil group means "the user opted into it". Tests cover four cases: value group with explicit flags, value group with defaults applied, pointer group allocated when an inner flag is set, pointer group left nil when nothing is set. Verified: go build ./..., go test ./shortcuts/common/..., and `go run golang.org/x/tools/cmd/deadcode@v0.31.0 -test ./...` against HEAD vs origin/main yields zero new dead code. This closes the previously-documented limitation that group sub-structs had to be nested inside an OneOf bucket to actually bind values.	2026-05-28 15:58:54 +08:00
shanglei	8d8acb8252	test(shortcuts): cover TypedShortcut descriptors, drop unused Mount After reverting the im pilot in `ccf654d3` the TypedShortcut framework has no production caller, so the CI deadcode check flagged five methods as newly unreachable. Resolve without breaking the framework: - Remove TypedShortcut.Mount — a convenience wrapper over MountWithContext with zero callers (production OR test); not part of any interface contract. - Add focused unit tests for the four ShortcutDescriptor methods required by the Mountable contract — GetAuthTypes plus ScopesForIdentity / ConditionalScopesForIdentity / DeclaredScopesForIdentity, with table-driven coverage of the user / bot / fallback / dedupe branches. Locally verified: go run golang.org/x/tools/cmd/deadcode@v0.31.0 -test ./... against HEAD vs origin/main yields an empty diff.	2026-05-28 10:57:39 +08:00
shanglei	ccf654d3f0	revert(shortcuts): drop im messages-send typed pilot, keep framework The im +messages-send TypedShortcut pilot was exploratory only; revert it while retaining the TypedShortcut framework for future migrations. - shortcuts/im: restored to pre-pilot state (im_messages_send.go back to legacy common.Shortcut; deleted protocol.go, protocol_test.go, im_messages_send_test.go; shortcuts.go drops TypedShortcuts()). - shortcuts/register.go: removed addTyped(im.TypedShortcuts()) wiring and the now-unused addTyped helper; legacy addLegacy + Mountable dispatch retained. - shortcuts/common, argstype, errs subtypes, cmd/auth adapters: kept. - framework doc comments: replaced examples referencing the removed pilot types (MessageTarget/MessageContent/VideoContent/RawContent) with neutral descriptions; noted no typed shortcut is registered today. Framework now has no production caller. Verified: go build ./... and go test ./shortcuts/... ./errs/... ./cmd/auth/... ./cmd/ all pass.	2026-05-27 18:16:17 +08:00
shanglei	ad4368ed2a	test(shortcuts): strengthen auth-type assertion + cover binder paths - im_messages_send_test: assert auth-type set membership (reject duplicates / missing members) per CodeRabbit review - binder_coverage_test: cover bindMaybe, bindBuckets / bindBucketInner, bucketLeafValue, runNormalize / asString, checkGroup, checkEnumAndRequired, and group recursion in runValidateValue — lifts patch coverage over the 60% gate	2026-05-27 15:32:58 +08:00
shanglei	a07239b923	feat(shortcuts): introduce TypedShortcut framework + im send pilot Add strongly-typed shortcut protocol that coexists with the legacy common.Shortcut. The new common.TypedShortcut[T] is a generic outer wrapper backed by a reflect-driven binder; both legacy and typed shortcuts satisfy a new common.Mountable / common.ShortcutDescriptor interface pair so register.go can dispatch either through the same pipeline. Framework (shortcuts/common): - protocol.go — Mountable / ShortcutDescriptor / OneOfMarker / Validatable / Normalizable[T] / ArgsValidator / Maybe[T] / HelpExample - binder.go — reflect Args walk, intra-Args flag-tag uniqueness panic, cobra flag registration, bindFlags + bindMaybe, runNormalize (via MethodByName dispatch — Normalizable[T] can't be type-asserted through a non-generic interface), runValidateValue, runFrameworkRules for required / enum / OneOf / group - typed_shortcut.go — TypedShortcut[T] struct, 8 descriptor methods, mountTyped adapter that synthesizes a legacy Shortcut shell and reuses runShortcut verbatim (identity / scopes / @file / stdin / jq / dry-run / high-risk gate) - typed_help.go — sectioned --help (CHOOSE ONE / OPTIONAL / EXAMPLES) with cmdutil.GetRisk/GetTips passthrough so typed shortcuts keep the Risk: and Tips: blocks - runner.go — typedArgs lifecycle slot on RuntimeContext - types.go — 5 GetX accessors on Shortcut so legacy shortcuts satisfy ShortcutDescriptor alongside the existing pointer-receiver scope methods Typed primitives (shortcuts/common/argstype): - ChatID / UserOpenID / UserOpenIDList — prefix-validated identifiers - SafePath — cwd-relative, rejects absolute paths and ".." segments - MediaInput — tri-state (URL bypass / img_xxx-file_xxx key bypass / SafePath delegation) - SpreadsheetRef — Normalize extracts shtcn token from feishu URLs Error contract (errs): - 3 new Subtype constants: shortcut_oneof_missing / shortcut_oneof_multiple / shortcut_group_incomplete. Per-field failures (required / enum / typed primitive format) reuse the existing SubtypeInvalidArgument so no new error type is introduced. Registry refactor (shortcuts + cmd): - AllShortcuts() now returns []common.ShortcutDescriptor; legacy shortcuts are boxed as Shortcut (pointer required for the pointer-receiver scope methods), typed shortcuts boxed as Mountable - Register dispatches via the Mountable interface - cmd/auth/login, cmd/auth/login_interactive, cmd/error_auth_hint, cmd/diagnose_scope_test, shortcuts/register_test (shortcuts.json generator) updated to read through GetService / GetCommand / GetAuthTypes / GetDescription / DeclaredScopesForIdentity - shortcutSupportsIdentity helpers accept ShortcutDescriptor Pilot (shortcuts/im): - protocol.go — MessageTarget / MessageContent (with seven content variants) / VideoContent (paired video + cover) / RawContent (--content with explicit msg-type, validates JSON in ValidateValue) - im_messages_send.go — migrated to TypedShortcut[*ImMessagesSendArgs]. Inline Validate closure is replaced by framework-derived checks (OneOf target, OneOf content, VideoContent group, typed-primitive formats, RawContent JSON). Helpers (resolveMediaContent, wrapMarkdownAsPostForDryRun, normalizeAtMentions, etc.) reused verbatim; only the field-access pattern changes from runtime.Str("x") to args.X. - shortcuts.go — new TypedShortcuts() exporter; ImMessagesSend removed from the legacy Shortcuts() slice so it is not double-mounted - register.go wires addTyped(im.TypedShortcuts()) into init Known follow-ups: - runFrameworkRules and bindFlags do not recurse into OneOf bucket / group sub-structs; im messages-send compensates with a local bindMessagesSendArgs + validateVideoGroup. Generalizing the binder to recurse will let future migrations drop the local shim. - common.ValidateChatID / common.ValidateUserID become redundant once all legacy shortcuts that call them migrate; can be retired with the last legacy caller. Refs: docs/superpowers/specs/2026-05-26-shortcut-protocol-design.md	2026-05-27 11:32:06 +08:00