chore: split slides creator skill

Change-Id: I2d3d47ef999a88b9aa8c22147e5a78d469889e5f

CHANGELOG.md

@@ -2,6 +2,23 @@
 
 All notable changes to this project will be documented in this file.
 
+## [v1.0.29] - 2026-05-12
+
+### Features
+
+- **vc**: Add agent meeting join, leave, and events shortcuts (#824)
+- **mail**: Add unknown-flag fuzzy match for `lark-cli mail` commands (#806)
+- **whiteboard**: Pin `whiteboard-cli` to `v0.2.11` in `lark-whiteboard` skill (#850)
+
+### Bug Fixes
+
+- Silence misleading "skills not installed" startup notice (#801)
+
+### Documentation
+
+- **base**: Refine data analysis SOP wording (#784, #849)
+- Update README capability descriptions (#793)
+
 ## [v1.0.28] - 2026-05-11
 
 ### Features
@@ -659,6 +676,7 @@ Bundled AI agent skills for intelligent assistance:
 - Bilingual documentation (English & Chinese).
 - CI/CD pipelines: linting, testing, coverage reporting, and automated releases.
 
+[v1.0.29]: https://github.com/larksuite/cli/releases/tag/v1.0.29
 [v1.0.28]: https://github.com/larksuite/cli/releases/tag/v1.0.28
 [v1.0.27]: https://github.com/larksuite/cli/releases/tag/v1.0.27
 [v1.0.26]: https://github.com/larksuite/cli/releases/tag/v1.0.26

README.md

@@ -6,14 +6,14 @@
 
 [中文版](./README.zh.md) | [English](./README.md)
 
-The official [Lark/Feishu](https://www.larksuite.com/) CLI tool, maintained by the [larksuite](https://github.com/larksuite) team — built for humans and AI Agents. Covers core business domains including Messenger, Docs, Base, Sheets, Slides, Calendar, Mail, Tasks, Meetings, Markdown, and more, with 200+ commands and 24 AI Agent [Skills](./skills/).
+The official [Lark/Feishu](https://www.larksuite.com/) CLI tool, maintained by the [larksuite](https://github.com/larksuite) team — built for humans and AI Agents. Covers core business domains including Messenger, Docs, Base, Sheets, Slides, Calendar, Mail, Tasks, Meetings, Markdown, and more, with 200+ commands and 25 AI Agent [Skills](./skills/).
 
 [Install](#installation--quick-start) · [AI Agent Skills](#agent-skills) · [Auth](#authentication) · [Commands](#three-layer-command-system) · [Advanced](#advanced-usage) · [Security](#security--risk-warnings-read-before-use) · [Contributing](#contributing)
 
 ## Why lark-cli?
 
-- **Agent-Native Design** — 24 structured [Skills](./skills/) out of the box, compatible with popular AI tools — Agents can operate Lark with zero extra setup
-- **Wide Coverage** — 17 business domains, 200+ curated commands, 24 AI Agent [Skills](./skills/)
+- **Agent-Native Design** — 25 structured [Skills](./skills/) out of the box, compatible with popular AI tools — Agents can operate Lark with zero extra setup
+- **Wide Coverage** — 17 business domains, 200+ curated commands, 25 AI Agent [Skills](./skills/)
 - **AI-Friendly & Optimized** — Every command is tested with real Agents, featuring concise parameters, smart defaults, and structured output to maximize Agent call success rates
 - **Open Source, Zero Barriers** — MIT license, ready to use, just `npm install`
 - **Up and Running in 3 Minutes** — One-click app creation, interactive login, from install to first API call in just 3 steps
@@ -24,7 +24,7 @@ The official [Lark/Feishu](https://www.larksuite.com/) CLI tool, maintained by t
 
 | Category      | Capabilities                                                                                                                      |
 | ------------- |-----------------------------------------------------------------------------------------------------------------------------------|
-| 📅 Calendar   | View agenda, create events, invite attendees, check free/busy status, time suggestions                                            |
+| 📅 Calendar   | View, create and update events, invite attendees, find meeting rooms, RSVP to invitations, check free/busy & time suggestions     |
 | 💬 Messenger  | Send/reply messages, create and manage group chats, view chat history & threads, search messages, download media                  |
 | 📄 Docs       | Create, read, update, and search documents, read/write media & whiteboards                                                        |
 | 📁 Drive      | Upload and download files, search docs & wiki, manage comments                                                                    |
@@ -36,7 +36,7 @@ The official [Lark/Feishu](https://www.larksuite.com/) CLI tool, maintained by t
 | 📚 Wiki       | Create and manage knowledge spaces, nodes, and documents                                                                          |
 | 👤 Contact    | Search users by name/email/phone, get user profiles                                                                               |
 | 📧 Mail       | Browse, search, read emails, send, reply, forward, manage drafts, watch new mail                                                  |
-| 🎥 Meetings   | Search meeting records, query meeting minutes & recordings                                                                        |
+| 🎥 Meetings   | Search meeting records, query meeting minutes artifacts and recordings                                                            |
 | 🕐 Attendance | Query personal attendance check-in records                                                                                        |
 | ✍️ 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.                                  |
@@ -136,13 +136,14 @@ lark-cli auth status
 | Skill                           | Description                                                                                                    |
 | ------------------------------- |----------------------------------------------------------------------------------------------------------------|
 | `lark-shared`                   | App config, auth login, identity switching, scope management, security rules (auto-loaded by all other skills) |
-| `lark-calendar`                 | Calendar events, agenda view, free/busy queries, time suggestions                                              |
+| `lark-calendar`                 | Calendar events (create/update), agenda view, free/busy queries, time suggestions, room finding, RSVP replies  |
 | `lark-im`                       | Send/reply messages, group chat management, message search, upload/download images & files, reactions          |
 | `lark-doc`                      | Create, read, update, search documents (Markdown-based)                                                        |
 | `lark-drive`                    | Upload, download files, manage permissions & comments                                                          |
 | `lark-markdown`                 | Create, fetch, and overwrite Drive-native Markdown files                                                       |
 | `lark-sheets`                   | Create, read, write, append, find, export spreadsheets                                                         |
-| `lark-slides`                   | Create and manage presentations, read presentation content, and add or remove slides                          |
+| `lark-slides-creator`           | Create polished presentations with planning, design, asset, template, and validation workflows                |
+| `lark-slides`                   | Low-level Slides XML/API read/write operations                                                               |
 | `lark-base`                     | Tables, fields, records, views, dashboards, data aggregation & analytics                                       |
 | `lark-task`                     | Tasks, task lists, subtasks, reminders, member assignment                                                      |
 | `lark-mail`                     | Browse, search, read emails, send, reply, forward, draft management, watch new mail                            |
@@ -151,7 +152,7 @@ lark-cli auth status
 | `lark-event`                    | Real-time event subscriptions (WebSocket), regex routing & agent-friendly format                               |
 | `lark-vc`                       | Search meeting records, query meeting minutes (summary, todos, transcript)                                     |
 | `lark-whiteboard`               | Whiteboard/chart DSL rendering                                                                                 |
-| `lark-minutes`                  | Minutes metadata & AI artifacts (summary, todos, chapters)                                                     |
+| `lark-minutes`                  | Minutes metadata & AI artifacts (summary, todos, chapters); upload audio/video to create minutes, download media |
 | `lark-openapi-explorer`         | Explore underlying APIs from official docs                                                                     |
 | `lark-skill-maker`              | Custom skill creation framework                                                                                |
 | `lark-attendance`               | Query personal attendance check-in records                                                                     |

README.zh.md

@@ -24,7 +24,7 @@
 
 | 类别        | 能力                                         |
 | ------------- |--------------------------------------------|
-| 📅 日历     | 查看日程、创建日程、邀请参会人、查询忙闲状态、时间建议                |
+| 📅 日历     | 查看、创建和更新日程，邀请参会人、查找会议室、回复日程邀请、查询忙闲与时间建议        |
 | 💬 即时通讯 | 发送/回复消息、创建和管理群聊、查看聊天记录与话题、搜索消息、下载媒体文件      |
 | 📄 云文档   | 创建、读取、更新文档、搜索文档、读写素材与画板                    |
 | 📁 云空间   | 上传和下载文件、搜索文档与知识库、管理评论                      |
@@ -36,7 +36,7 @@
 | 📚 知识库   | 创建和管理知识空间、节点和文档                            |
 | 👤 通讯录   | 按姓名/邮箱/手机号搜索用户、获取用户信息                      |
 | 📧 邮箱     | 浏览、搜索、阅读邮件，发送、回复、转发邮件，管理草稿，监听新邮件           |
-| 🎥 视频会议 | 搜索会议记录、查询会议纪要与录制                           |
+| 🎥 视频会议 | 搜索会议记录、查询会议纪要产物与会议录制                        |
 | 🕐 考勤打卡 | 查询个人考勤打卡记录                                |
 | ✍️ 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        |
@@ -137,7 +137,7 @@ lark-cli auth status
 | Skill                           | 说明                                        |
 | --------------------------------- |-------------------------------------------|
 | `lark-shared`                   | 应用配置、认证登录、身份切换、权限管理、安全规则（所有其他 skill 自动加载） |
-| `lark-calendar`                 | 日历日程、议程查看、忙闲查询、时间建议                       |
+| `lark-calendar`                 | 日历日程（创建/更新）、议程查看、忙闲查询、时间建议、会议室查找、回复邀请       |
 | `lark-im`                       | 发送/回复消息、群聊管理、消息搜索、上传下载图片与文件、表情回复          |
 | `lark-doc`                      | 创建、读取、更新、搜索文档（基于 Markdown）                |
 | `lark-drive`                    | 上传、下载文件，管理权限与评论                           |
@@ -152,7 +152,7 @@ lark-cli auth status
 | `lark-event`                    | 实时事件订阅（WebSocket），支持正则路由与 Agent 友好格式      |
 | `lark-vc`                       | 搜索会议记录、查询会议纪要产物（总结、待办、逐字稿）                |
 | `lark-whiteboard`               | 画板/图表 DSL 渲染                              |
-| `lark-minutes`                  | 妙记元数据与 AI 产物（总结、待办、章节）                    |
+| `lark-minutes`                  | 妙记元数据与 AI 产物（总结、待办、章节），上传音视频生成妙记，下载音视频文件 |
 | `lark-openapi-explorer`         | 从官方文档探索底层 API                             |
 | `lark-skill-maker`              | 自定义 skill 创建框架                            |
 | `lark-attendance`               | 查询个人考勤打卡记录                                |

cmd/auth/login.go

@@ -232,7 +232,7 @@ func authLoginRun(opts *LoginOptions) error {
 			"verification_url": authResp.VerificationUriComplete,
 			"device_code":      authResp.DeviceCode,
 			"expires_in":       authResp.ExpiresIn,
-			"hint":             fmt.Sprintf("Show verification_url to user, then immediately execute: lark-cli auth login --device-code %s (blocks until authorized or timeout). Do not instruct the user to run this command themselves.", authResp.DeviceCode),
+			"hint":             fmt.Sprintf("Show verification_url to the user exactly as returned by the CLI and treat it as an opaque string. Do not URL-encode or decode it, do not normalize or rewrite it, do not add %%20, spaces, or punctuation, and do not wrap it as Markdown link text; prefer a fenced code block containing only the raw URL. Then immediately execute: lark-cli auth login --device-code %s (blocks until authorized or timeout). Do not instruct the user to run this command themselves.", authResp.DeviceCode),
 		}
 		encoder := json.NewEncoder(f.IOStreams.Out)
 		encoder.SetEscapeHTML(false)

cmd/auth/login_messages.go

@@ -59,7 +59,7 @@ var loginMsgZh = &loginMsg{
 
 	OpenURL:            "在浏览器中打开以下链接进行认证:\n\n",
 	WaitingAuth:        "等待用户授权...",
-	AgentTimeoutHint:   "[AI agent] 此命令最长阻塞约 10 分钟，等待用户在浏览器内完成授权。请确保 runner 的 timeout ≥ 600s；如不支持长 timeout，请改用 `lark-cli auth login --no-wait --json` 拿到 device_code 后再用 `lark-cli auth login --device-code <code>` 续上轮询，**不要短 timeout 反复重试**——每次重启会作废上一轮的 device code，导致用户授权的链接失效。",
+	AgentTimeoutHint:   "[AI agent] 此命令最长阻塞约 10 分钟，等待用户在浏览器内完成授权。请确保 runner 的 timeout >= 600s；如不支持长 timeout，请改用 `lark-cli auth login --no-wait --json` 拿到 device_code 后再用 `lark-cli auth login --device-code <code>` 续上轮询。**不要短 timeout 反复重试**，每次重启会作废上一轮的 device code，导致用户授权链接失效。向用户展示授权链接时，必须逐字原样转发 CLI 返回的 URL，把它视为不可修改的 opaque string；不要做 URL 编码或解码，不要补 `%20`、空格或标点，不要改写成 Markdown 链接，建议用只包含该 URL 的代码块单独输出。",
 	AuthSuccess:        "已收到授权确认，正在获取用户信息并校验授权结果...",
 	LoginSuccess:       "授权成功! 用户: %s (%s)",
 	AuthorizedUser:     "当前授权账号: %s (%s)",
@@ -95,7 +95,7 @@ var loginMsgEn = &loginMsg{
 
 	OpenURL:            "Open this URL in your browser to authenticate:\n\n",
 	WaitingAuth:        "Waiting for user authorization...",
-	AgentTimeoutHint:   "[AI agent] This command blocks for up to ~10 minutes while waiting for the user to authorize in their browser. Make sure your runner's timeout is ≥ 600s. If long timeouts are not supported, use `lark-cli auth login --no-wait --json` to get a device_code, then `lark-cli auth login --device-code <code>` to resume polling. **Do NOT retry with a short timeout** — each restart invalidates the previous device code, so any URL the user already authorized becomes useless.",
+	AgentTimeoutHint:   "[AI agent] This command blocks for up to ~10 minutes while waiting for the user to authorize in their browser. Make sure your runner's timeout is >= 600s. If long timeouts are not supported, use `lark-cli auth login --no-wait --json` to get a device_code, then `lark-cli auth login --device-code <code>` to resume polling. **Do NOT retry with a short timeout**; each restart invalidates the previous device code and makes the earlier authorization URL useless. When showing the authorization URL to the user, copy the CLI-returned URL exactly as-is and treat it as an opaque string. Do not URL-encode or decode it, do not add `%20`, spaces, or punctuation, do not rewrite it as Markdown link text, and prefer a fenced code block containing only the raw URL.",
 	AuthSuccess:        "Authorization confirmed, fetching user info and validating granted scopes...",
 	LoginSuccess:       "Authorization successful! User: %s (%s)",
 	AuthorizedUser:     "Authorized account: %s (%s)",

cmd/auth/login_test.go

@@ -879,6 +879,57 @@ func TestAuthLoginRun_JSONWriteFailure_NoWaitReturnsWriterError(t *testing.T) {
 	}
 }
 
+func TestAuthLoginRun_NoWaitJSONHintIncludesRawURLGuidance(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
+		ProfileName: "default",
+		AppID:       "cli_test",
+		AppSecret:   "secret",
+		Brand:       core.BrandFeishu,
+	})
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    larkauth.PathDeviceAuthorization,
+		Body: map[string]interface{}{
+			"device_code":               "device-code",
+			"user_code":                 "user-code",
+			"verification_uri":          "https://example.com/verify",
+			"verification_uri_complete": "https://example.com/verify?code=123",
+			"expires_in":                240,
+			"interval":                  5,
+		},
+	})
+
+	err := authLoginRun(&LoginOptions{
+		Factory: f,
+		Ctx:     context.Background(),
+		Scope:   "im:message:send",
+		NoWait:  true,
+	})
+	if err != nil {
+		t.Fatalf("authLoginRun() error = %v", err)
+	}
+
+	dec := json.NewDecoder(strings.NewReader(stdout.String()))
+	var data map[string]interface{}
+	if err := dec.Decode(&data); err != nil {
+		t.Fatalf("Decode(stdout first event) error = %v, stdout=%q", err, stdout.String())
+	}
+	hint, _ := data["hint"].(string)
+	for _, want := range []string{
+		"exactly as returned by the CLI",
+		"opaque string",
+		"Do not URL-encode or decode it",
+		"do not add %20, spaces, or punctuation",
+		"do not wrap it as Markdown link text",
+		"fenced code block containing only the raw URL",
+	} {
+		if !strings.Contains(hint, want) {
+			t.Fatalf("hint missing %q, got:\n%s", want, hint)
+		}
+	}
+}
+
 func TestAuthLoginRun_JSONWriteFailure_DeviceAuthorizationReturnsWriterError(t *testing.T) {
 	f, _, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
 		ProfileName: "default",
@@ -917,6 +968,60 @@ func TestAuthLoginRun_JSONWriteFailure_DeviceAuthorizationReturnsWriterError(t *
 	}
 }
 
+func TestAuthLoginRun_JSONDeviceAuthorizationAgentHintIncludesRawURLGuidance(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, &core.CliConfig{
+		ProfileName: "default",
+		AppID:       "cli_test",
+		AppSecret:   "secret",
+		Brand:       core.BrandFeishu,
+	})
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    larkauth.PathDeviceAuthorization,
+		Body: map[string]interface{}{
+			"device_code":               "device-code",
+			"user_code":                 "user-code",
+			"verification_uri":          "https://example.com/verify",
+			"verification_uri_complete": "https://example.com/verify?code=123",
+			"expires_in":                240,
+			"interval":                  5,
+		},
+	})
+
+	ctx, cancel := context.WithCancel(context.Background())
+	cancel()
+	err := authLoginRun(&LoginOptions{
+		Factory: f,
+		Ctx:     ctx,
+		Scope:   "im:message:send",
+		JSON:    true,
+	})
+	if err == nil {
+		t.Fatal("expected error from cancelled context")
+	}
+
+	dec := json.NewDecoder(strings.NewReader(stdout.String()))
+	var data map[string]interface{}
+	if err := dec.Decode(&data); err != nil {
+		t.Fatalf("Decode(stdout first event) error = %v, stdout=%q", err, stdout.String())
+	}
+	hint, _ := data["agent_hint"].(string)
+	for _, want := range []string{
+		"timeout >= 600s",
+		"逐字原样转发 CLI 返回的 URL",
+		"opaque string",
+		"不要做 URL 编码或解码",
+		"不要补 `%20`、空格或标点",
+		"不要改写成 Markdown 链接",
+		"只包含该 URL 的代码块单独输出",
+	} {
+		if !strings.Contains(hint, want) {
+			t.Fatalf("agent_hint missing %q, got:\n%s", want, hint)
+		}
+	}
+}
+
 func TestGetDomainMetadata_ExcludesEvent(t *testing.T) {
 	domains := getDomainMetadata("zh")
 	for _, dm := range domains {

cmd/root_integration_test.go

@@ -504,10 +504,12 @@ func TestIntegration_Shortcut_BusinessError_OutputsEnvelope(t *testing.T) {
 	})
 }
 
-// TestSetupNotices_ColdStart verifies that when no skills stamp exists,
-// the composed PendingNotice provider includes a "skills" key with an
-// empty Current and the cold-start message.
-func TestSetupNotices_ColdStart(t *testing.T) {
+// TestSetupNotices_ColdStart_NoNotice verifies that a missing stamp
+// produces no skills key in the composed notice. Users who installed
+// skills via `npx skills add` (no stamp) must not see the misleading
+// "not installed" notice — only `lark-cli update` users opt into the
+// drift tracker.
+func TestSetupNotices_ColdStart_NoNotice(t *testing.T) {
 	clearNoticeEnv(t)
 	dir := t.TempDir()
 	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", dir)
@@ -530,17 +532,10 @@ func TestSetupNotices_ColdStart(t *testing.T) {
 
 	notice := output.GetNotice()
 	if notice == nil {
-		t.Fatal("GetNotice() = nil, want non-nil for cold start")
+		return // expected — no pending notices at all
 	}
-	skills, ok := notice["skills"].(map[string]interface{})
-	if !ok {
-		t.Fatalf("notice.skills missing, got %+v", notice)
-	}
-	if skills["current"] != "" || skills["target"] != "1.0.21" {
-		t.Errorf("notice.skills = %+v, want {current:\"\", target:\"1.0.21\"}", skills)
-	}
-	if msg, _ := skills["message"].(string); msg != "lark-cli skills not installed, run: lark-cli update" {
-		t.Errorf("notice.skills.message = %q, want cold-start message", msg)
+	if _, ok := notice["skills"]; ok {
+		t.Errorf("notice.skills present in cold-start state, want absent: %+v", notice)
 	}
 }
 

cmd/update/update_test.go

@@ -168,6 +168,11 @@ func TestUpdateManual_Human(t *testing.T) {
 }
 
 func TestUpdateNpm_JSON(t *testing.T) {
+	// Isolate config dir: this test mocks fetchLatest="2.0.0" and lets
+	// runSkillsAndStamp → WriteStamp succeed, which without isolation would
+	// clobber the real ~/.lark-cli/skills.stamp with "2.0.0".
+	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
+
 	f, stdout, _ := newTestFactory(t)
 	cmd := NewCmdUpdate(f)
 	cmd.SetArgs([]string{"--json"})
@@ -195,6 +200,9 @@ func TestUpdateNpm_JSON(t *testing.T) {
 }
 
 func TestUpdateNpm_Human(t *testing.T) {
+	// Same isolation as TestUpdateNpm_JSON — see comment there.
+	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
+
 	f, _, stderr := newTestFactory(t)
 	cmd := NewCmdUpdate(f)
 	cmd.SetArgs([]string{})
@@ -222,6 +230,9 @@ func TestUpdateNpm_Human(t *testing.T) {
 }
 
 func TestUpdateForce_JSON(t *testing.T) {
+	// Same stamp-isolation rationale as TestUpdateNpm_JSON.
+	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
+
 	f, stdout, _ := newTestFactory(t)
 	cmd := NewCmdUpdate(f)
 	cmd.SetArgs([]string{"--force", "--json"})
@@ -312,6 +323,9 @@ func TestUpdateInvalidVersion_JSON(t *testing.T) {
 }
 
 func TestUpdateDevVersion_JSON(t *testing.T) {
+	// Same stamp-isolation rationale as TestUpdateNpm_JSON.
+	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
+
 	f, stdout, _ := newTestFactory(t)
 	cmd := NewCmdUpdate(f)
 	cmd.SetArgs([]string{"--json"})
@@ -629,6 +643,9 @@ func TestPermissionHint(t *testing.T) {
 
 func TestUpdateWindows_NpmSuccess_JSON(t *testing.T) {
 	// With the rename trick, Windows npm installs can now auto-update.
+	// Same stamp-isolation rationale as TestUpdateNpm_JSON.
+	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
+
 	f, stdout, _ := newTestFactory(t)
 	cmd := NewCmdUpdate(f)
 	cmd.SetArgs([]string{"--json"})

internal/binding/audit.go

@@ -65,7 +65,11 @@ func AssertSecurePath(params AuditParams) (string, error) {
 }
 
 // requireAbsolutePath rejects relative paths; relative paths would depend on
-// the process cwd and defeat the point of a static audit.
+// the process cwd and defeat the point of a static audit. Shell-style
+// shortcuts like `~` are home-relative, not cwd-relative — they are an
+// orthogonal concern and the audit is intentionally Go-stdlib strict here.
+// Callers that accept user-authored config (e.g. resolveFileRef) must
+// pre-resolve any such shortcuts before passing the path in.
 func requireAbsolutePath(target, label string) error {
 	if !filepath.IsAbs(target) {
 		return fmt.Errorf("%s: path must be absolute, got %q", label, target)

internal/binding/secret_resolve_file.go

@@ -23,9 +23,19 @@ func resolveFileRef(ref *SecretRef, pc *ProviderConfig) (string, error) {
 		return "", fmt.Errorf("file provider path is empty")
 	}
 
+	// OpenClaw preserves user-authored `~/...` paths verbatim on disk for
+	// portability and resolves them at read time. lark-cli reads the file
+	// raw, so we mirror that resolution here before the audit — otherwise
+	// an unambiguous home-relative path would be rejected by
+	// requireAbsolutePath, which is meant to guard against cwd-relative
+	// paths (a different concern). expandTildePath honours OPENCLAW_HOME so
+	// a tilde inside an OPENCLAW_HOME-overridden config resolves to the
+	// same absolute path OpenClaw itself would have used.
+	targetPath := expandTildePath(pc.Path)
+
 	// Security audit on file path
 	securePath, err := AssertSecurePath(AuditParams{
-		TargetPath:            pc.Path,
+		TargetPath:            targetPath,
 		Label:                 "secrets.providers file path",
 		TrustedDirs:           pc.TrustedDirs,
 		AllowInsecurePath:     pc.AllowInsecurePath,

internal/binding/secret_resolve_file_test.go

@@ -6,6 +6,7 @@ package binding
 import (
 	"os"
 	"path/filepath"
+	"strings"
 	"testing"
 )
 
@@ -230,3 +231,88 @@ func TestResolveFileRef_ExceedsMaxBytes(t *testing.T) {
 		t.Errorf("error = %q, want %q", err.Error(), want)
 	}
 }
+
+// TestResolveFileRef_TildePath_SingleValue is the end-to-end smoke test
+// for the fix: a singleValue file provider with a ~/-relative path
+// resolves correctly through resolveFileRef. Before this PR the audit
+// would reject the path as "must be absolute".
+func TestResolveFileRef_TildePath_SingleValue(t *testing.T) {
+	dir := t.TempDir()
+	setFakeOSHome(t, dir)
+	t.Setenv("OPENCLAW_HOME", "")
+
+	p := filepath.Join(dir, "secret.txt")
+	if err := os.WriteFile(p, []byte("tilde_secret\n"), 0o600); err != nil {
+		t.Fatalf("write temp file: %v", err)
+	}
+
+	ref := &SecretRef{Source: "file", ID: SingleValueFileRefID}
+	pc := &ProviderConfig{
+		Source:            "file",
+		Path:              "~/secret.txt",
+		Mode:              "singleValue",
+		AllowInsecurePath: true,
+	}
+
+	got, err := resolveFileRef(ref, pc)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if got != "tilde_secret" {
+		t.Errorf("got %q, want %q", got, "tilde_secret")
+	}
+}
+
+// TestResolveFileRef_RelativePath_StillRejected guards the absolute-path
+// audit: cwd-relative input must still be rejected even though tilde was
+// loosened. Catches regressions if expandTildePath is ever widened to
+// also expand "./..." (which would weaken the audit's invariant).
+func TestResolveFileRef_RelativePath_StillRejected(t *testing.T) {
+	ref := &SecretRef{Source: "file", ID: SingleValueFileRefID}
+	pc := &ProviderConfig{
+		Source:            "file",
+		Path:              "relative/secret.txt",
+		Mode:              "singleValue",
+		AllowInsecurePath: true,
+	}
+
+	_, err := resolveFileRef(ref, pc)
+	if err == nil {
+		t.Fatal("expected error for relative path, got nil")
+	}
+	wantSub := "path must be absolute"
+	if !strings.Contains(err.Error(), wantSub) {
+		t.Errorf("error = %q, want substring %q", err.Error(), wantSub)
+	}
+}
+
+// TestResolveFileRef_TildePath_JSONMode verifies the tilde-expansion
+// path works for json mode (where ref id is a JSON pointer) as well as
+// singleValue mode — the mechanism is mode-agnostic.
+func TestResolveFileRef_TildePath_JSONMode(t *testing.T) {
+	dir := t.TempDir()
+	setFakeOSHome(t, dir)
+	t.Setenv("OPENCLAW_HOME", "")
+
+	p := filepath.Join(dir, "secrets.json")
+	content := `{"providers":{"feishu":{"key":"json_via_tilde"}}}`
+	if err := os.WriteFile(p, []byte(content), 0o600); err != nil {
+		t.Fatalf("write temp file: %v", err)
+	}
+
+	ref := &SecretRef{Source: "file", ID: "/providers/feishu/key"}
+	pc := &ProviderConfig{
+		Source:            "file",
+		Path:              "~/secrets.json",
+		Mode:              "json",
+		AllowInsecurePath: true,
+	}
+
+	got, err := resolveFileRef(ref, pc)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if got != "json_via_tilde" {
+		t.Errorf("got %q, want %q", got, "json_via_tilde")
+	}
+}

internal/binding/tilde.go

@@ -0,0 +1,180 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package binding
+
+import (
+	"os"
+	"os/user"
+	"path/filepath"
+	"strings"
+
+	"github.com/larksuite/cli/internal/vfs"
+)
+
+// hasTildePrefix reports whether s begins with `~` followed by end-of-string,
+// `/`, or `\` — the form OpenClaw treats as home-relative.
+func hasTildePrefix(s string) bool {
+	if s == "" || s[0] != '~' {
+		return false
+	}
+	if len(s) == 1 {
+		return true
+	}
+	return s[1] == '/' || s[1] == '\\'
+}
+
+// joinTildeSuffix expands a tilde-prefixed string against a resolved home
+// directory. Replaces only the leading `~` so the original separator
+// (forward or back slash) and suffix bytes are kept verbatim, matching
+// OpenClaw's `input.replace(/^~(?=$|[\\/])/, home)` semantics rather than
+// going through filepath.Join (which would silently drop a literal `\` on
+// POSIX). filepath.Clean is applied so `..` and duplicate separators are
+// collapsed in the same way Node's path.resolve does on each platform.
+//
+// Caller must ensure hasTildePrefix(s) is true and home is non-empty.
+func joinTildeSuffix(s, home string) string {
+	if len(s) == 1 {
+		return home
+	}
+	return filepath.Clean(home + s[1:])
+}
+
+// normalizeSentinel applies OpenClaw's normalize() helper to a single
+// string: trims whitespace and treats the JS-flavoured literals
+// "undefined" / "null" (along with empty/whitespace-only) as unset.
+func normalizeSentinel(v string) string {
+	v = strings.TrimSpace(v)
+	if v == "undefined" || v == "null" {
+		return ""
+	}
+	return v
+}
+
+// osHome returns the OS-level home directory by walking OpenClaw's
+// resolution chain: HOME → USERPROFILE → OS user database (getpwuid on
+// Unix / user32 on Windows, via os/user.Current). Each candidate is
+// passed through normalizeSentinel so sentinel literals and blank
+// strings fall through.
+//
+// Matches OpenClaw's resolveRawOsHomeDir env chain so the same tilde
+// resolves against the same home under mixed shell environments and
+// accidentally-stringified env values. Go's stdlib os.UserHomeDir on
+// Unix only re-reads HOME and gives up; Node's os.homedir() still
+// returns the account home via the user database, so the explicit
+// user.Current() step is what keeps OpenClaw-authored `~/...` working
+// in HOME-unset shells.
+//
+// Deliberate hybrid contract — neither a strict mirror of OpenClaw
+// nor a strict reject-on-missing:
+//
+//   - OpenClaw's final fallback is cwd (via resolveRequiredHomeDir →
+//     process.cwd()). We don't do that because requireAbsolutePath
+//     exists precisely to reject cwd-dependent paths; routing
+//     `~/secret` through cwd would defeat the audit invariant.
+//
+//   - We still go through user.Current() before giving up, even when
+//     HOME is a sentinel literal ("undefined" / "null") and
+//     USERPROFILE is unset. At that point OpenClaw would land on cwd,
+//     and a strict implementation would reject; user.Current() lands
+//     on the account home instead — cwd-independent and user-bound,
+//     so it satisfies the audit's safety goal while still letting
+//     ~/-authored configs resolve in a malformed-env shell.
+//
+//   - Only returns "" when the env chain AND user.Current() are all
+//     unresolvable, at which point the caller surfaces a clean
+//     "path must be absolute" error from the audit.
+func osHome() string {
+	if v := normalizeSentinel(os.Getenv("HOME")); v != "" {
+		return v
+	}
+	if v := normalizeSentinel(os.Getenv("USERPROFILE")); v != "" {
+		return v
+	}
+	if u, err := user.Current(); err == nil {
+		return normalizeSentinel(u.HomeDir)
+	}
+	return ""
+}
+
+// explicitOpenClawHome reads OPENCLAW_HOME with OpenClaw's normalize()
+// semantics applied.
+func explicitOpenClawHome() string {
+	return normalizeSentinel(os.Getenv("OPENCLAW_HOME"))
+}
+
+// absolutize returns p as an absolute path, resolving against the process
+// cwd when p is relative. Returns "" when the cwd cannot be resolved.
+// Wraps filepath.Abs semantics via vfs.Getwd because forbidigo bans
+// filepath.Abs inside internal/ packages.
+func absolutize(p string) string {
+	if p == "" {
+		return ""
+	}
+	if filepath.IsAbs(p) {
+		return filepath.Clean(p)
+	}
+	wd, err := vfs.Getwd()
+	if err != nil {
+		return ""
+	}
+	return filepath.Join(wd, p)
+}
+
+// openClawHome returns the home directory used to resolve `~`-relative paths
+// authored against OpenClaw's config. Closely mirrors OpenClaw's
+// home-resolution semantics so the same tilde resolves to the same
+// absolute path here as inside OpenClaw runtime under all normal
+// conditions.
+//
+// Resolution order:
+//  1. OPENCLAW_HOME env var, when set (sentinel-normalised).
+//  2. If OPENCLAW_HOME itself has a tilde prefix, expand it against the OS
+//     home (see osHome); the result is empty when the OS home is
+//     unresolvable.
+//  3. Otherwise fall back to the OS home.
+//
+// The returned path is absolute (relative OPENCLAW_HOME values are
+// absolutised against the process cwd, matching Node path.resolve in
+// OpenClaw's pipeline).
+//
+// Returns "" when no home can be resolved. This is a deliberate
+// divergence from OpenClaw, whose read pipeline would fall back to
+// cwd via resolveRequiredHomeDir — see osHome for the rationale.
+func openClawHome() string {
+	raw := explicitOpenClawHome()
+	switch {
+	case raw == "":
+		raw = osHome()
+	case hasTildePrefix(raw):
+		h := osHome()
+		if h == "" {
+			return ""
+		}
+		raw = joinTildeSuffix(raw, h)
+	}
+	return absolutize(raw)
+}
+
+// expandTildePath resolves a leading `~` or `~/...` prefix to OpenClaw's
+// effective home directory (see openClawHome).
+//
+// Returns the input unchanged when it lacks a tilde prefix or when
+// openClawHome cannot resolve a home directory. The latter case is a
+// deliberate divergence from OpenClaw, whose read pipeline falls back
+// to cwd — see osHome. Surfacing a "path must be absolute" error from
+// the audit is preferable to silently routing a user-authored
+// `~/secret` through cwd resolution.
+//
+// `~user` shell-style expansion is intentionally not supported (OpenClaw
+// does not support it either).
+func expandTildePath(p string) string {
+	if !hasTildePrefix(p) {
+		return p
+	}
+	home := openClawHome()
+	if home == "" {
+		return p
+	}
+	return joinTildeSuffix(p, home)
+}

internal/binding/tilde_test.go

@@ -0,0 +1,293 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package binding
+
+import (
+	"os"
+	"os/user"
+	"path/filepath"
+	"runtime"
+	"strings"
+	"testing"
+)
+
+// setFakeOSHome controls osHome's env-chain inputs (HOME and USERPROFILE)
+// in one call so tests stay deterministic across platforms. osHome reads
+// HOME first, then USERPROFILE, then user.Current(); setting only one of
+// the two leaves the test sensitive to whichever the runner happens to
+// have populated. Passing dir == "" disables both env entries so tests
+// can exercise the user.Current() fallback or no-home edge cases.
+func setFakeOSHome(t *testing.T, dir string) {
+	t.Helper()
+	t.Setenv("HOME", dir)
+	t.Setenv("USERPROFILE", dir)
+}
+
+// isolateRuntimeWrites parks the process cwd in a fresh TempDir for the
+// test's duration. Tests that set HOME to a sentinel literal trigger Go
+// runtime side effects — most visibly the telemetry subsystem, which
+// calls os.UserConfigDir() (= "$HOME/Library/Application Support" on
+// darwin) and happily writes through a relative result like
+// "undefined/Library/...". Without isolation those files land in the
+// package or repo dir and get accidentally staged. Chdir'ing into a
+// TempDir routes the noise into a path testing.T auto-cleans.
+func isolateRuntimeWrites(t *testing.T) {
+	t.Helper()
+	orig, err := os.Getwd()
+	if err != nil {
+		t.Fatalf("getwd: %v", err)
+	}
+	if err := os.Chdir(t.TempDir()); err != nil {
+		t.Fatalf("chdir: %v", err)
+	}
+	t.Cleanup(func() {
+		_ = os.Chdir(orig)
+	})
+}
+
+// TestOpenClawHome covers the openClawHome resolution table: empty /
+// sentinel OPENCLAW_HOME falls back to the OS home, explicit absolute
+// values are used verbatim (with whitespace trimmed), and tilde-prefixed
+// values recurse through the OS home.
+func TestOpenClawHome(t *testing.T) {
+	homeDir := t.TempDir()
+	explicit := t.TempDir()
+	setFakeOSHome(t, homeDir)
+
+	tests := []struct {
+		name        string
+		openclawEnv string
+		want        string
+	}{
+		{"unset falls back to OS home", "", homeDir},
+		{"undefined literal treated as unset", "undefined", homeDir},
+		{"null literal treated as unset", "null", homeDir},
+		{"whitespace-only treated as unset", "   ", homeDir},
+		{"explicit absolute path used verbatim", explicit, explicit},
+		{"explicit absolute path is trimmed", "  " + explicit + "  ", explicit},
+		{"bare tilde resolves to OS home", "~", homeDir},
+		{"tilde-prefixed value recurses through OS home", "~/custom", filepath.Join(homeDir, "custom")},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			t.Setenv("OPENCLAW_HOME", tc.openclawEnv)
+			got := openClawHome()
+			if got != tc.want {
+				t.Errorf("openClawHome() = %q, want %q", got, tc.want)
+			}
+		})
+	}
+}
+
+// TestOpenClawHome_RelativeIsAbsolutized confirms a relative
+// OPENCLAW_HOME is resolved against the process cwd, mirroring Node's
+// path.resolve behaviour in OpenClaw.
+func TestOpenClawHome_RelativeIsAbsolutized(t *testing.T) {
+	t.Setenv("OPENCLAW_HOME", filepath.FromSlash("relative/dir"))
+	got := openClawHome()
+
+	if !filepath.IsAbs(got) {
+		t.Fatalf("openClawHome() = %q, want absolute path", got)
+	}
+	wantSuffix := filepath.FromSlash("relative/dir")
+	if !strings.HasSuffix(got, wantSuffix) {
+		t.Errorf("openClawHome() = %q, want suffix %q", got, wantSuffix)
+	}
+}
+
+// TestOpenClawHome_FallsBackToUserDatabase pins osHome's final fallback
+// to the OS user database when HOME and USERPROFILE are both unset,
+// matching Node's os.homedir() (which uses getpwuid). Cwd-independent
+// and user-bound, so it does not conflict with the "no cwd fallback"
+// rule documented on osHome.
+func TestOpenClawHome_FallsBackToUserDatabase(t *testing.T) {
+	u, err := user.Current()
+	if err != nil || u.HomeDir == "" {
+		t.Skip("os/user.Current() unavailable on this runner")
+	}
+	setFakeOSHome(t, "")
+	t.Setenv("OPENCLAW_HOME", "")
+	got := openClawHome()
+	if got != u.HomeDir {
+		t.Errorf("openClawHome() = %q, want %q (account home from user.Current)", got, u.HomeDir)
+	}
+}
+
+// TestOpenClawHome_TildeOpenClawHomeUsesUserDatabaseFallback pins that
+// a tilde-form OPENCLAW_HOME ("~/custom") expands against the
+// user-database fallback when HOME and USERPROFILE are both unset.
+// Without the user.Current() step in osHome this would have failed
+// (returning "") and dropped the bind back to the audit's
+// "path must be absolute" error.
+func TestOpenClawHome_TildeOpenClawHomeUsesUserDatabaseFallback(t *testing.T) {
+	u, err := user.Current()
+	if err != nil || u.HomeDir == "" {
+		t.Skip("os/user.Current() unavailable on this runner")
+	}
+	setFakeOSHome(t, "")
+	t.Setenv("OPENCLAW_HOME", "~/custom")
+	got := openClawHome()
+	want := filepath.Join(u.HomeDir, "custom")
+	if got != want {
+		t.Errorf("openClawHome() = %q, want %q", got, want)
+	}
+}
+
+// TestExpandTildePath covers the full input grid for expandTildePath:
+// bare tilde, tilde-slash, tilde + suffix, nested suffix, plain absolute
+// and relative literals, and the intentionally-unchanged forms (~user,
+// ~foo) that OpenClaw does not expand either.
+func TestExpandTildePath(t *testing.T) {
+	fakeHome := t.TempDir()
+	absFixture := filepath.Join(fakeHome, "abs.json")
+	setFakeOSHome(t, fakeHome)
+	t.Setenv("OPENCLAW_HOME", "")
+
+	tests := []struct {
+		name string
+		in   string
+		want string
+	}{
+		{"empty", "", ""},
+		{"bare tilde", "~", fakeHome},
+		{"tilde slash", "~/", fakeHome},
+		{"tilde with file", "~/secret.json", filepath.Join(fakeHome, "secret.json")},
+		{"tilde with nested path", "~/.openclaw/secret.json", filepath.Join(fakeHome, ".openclaw/secret.json")},
+		{"absolute unchanged", absFixture, absFixture},
+		{"relative unchanged", "foo/bar", "foo/bar"},
+		{"dot relative unchanged", "../foo", "../foo"},
+		{"tilde user form unchanged", "~root/foo", "~root/foo"},
+		{"tilde without separator unchanged", "~foo", "~foo"},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			got := expandTildePath(tc.in)
+			if got != tc.want {
+				t.Errorf("expandTildePath(%q) = %q, want %q", tc.in, got, tc.want)
+			}
+		})
+	}
+}
+
+// TestExpandTildePath_RespectsOpenClawHome verifies that with
+// OPENCLAW_HOME set, tilde expansion uses that custom home rather than
+// the OS home — the integration-level invariant that closes the
+// internal inconsistency CodeX's first review flagged.
+func TestExpandTildePath_RespectsOpenClawHome(t *testing.T) {
+	homeDir := t.TempDir()
+	clawHome := t.TempDir()
+	setFakeOSHome(t, homeDir)
+	t.Setenv("OPENCLAW_HOME", clawHome)
+
+	got := expandTildePath("~/secret.json")
+	want := filepath.Join(clawHome, "secret.json")
+	if got != want {
+		t.Errorf("expandTildePath(%q) = %q, want %q (should use OPENCLAW_HOME)", "~/secret.json", got, want)
+	}
+	if got == filepath.Join(homeDir, "secret.json") {
+		t.Errorf("expandTildePath unexpectedly used OS home %q instead of OPENCLAW_HOME %q", homeDir, clawHome)
+	}
+}
+
+// TestExpandTildePath_FallsBackToUserDatabase is the end-to-end
+// equivalent of TestOpenClawHome_FallsBackToUserDatabase: with HOME and
+// USERPROFILE both unset, expandTildePath still resolves `~/foo` via
+// osHome's user.Current() step. Matches Node os.homedir() and keeps
+// OpenClaw-authored configs working in minimal-env shells.
+func TestExpandTildePath_FallsBackToUserDatabase(t *testing.T) {
+	u, err := user.Current()
+	if err != nil || u.HomeDir == "" {
+		t.Skip("os/user.Current() unavailable on this runner")
+	}
+	setFakeOSHome(t, "")
+	t.Setenv("OPENCLAW_HOME", "")
+	got := expandTildePath("~/foo")
+	want := filepath.Join(u.HomeDir, "foo")
+	if got != want {
+		t.Errorf("expandTildePath(~/foo) = %q, want %q", got, want)
+	}
+}
+
+// TestOpenClawHome_OSHomeNormalization pins OpenClaw's sentinel
+// normalisation on the env chain: the literals "undefined" / "null" /
+// blank-or-whitespace are all treated as unset, so a JS-flavoured
+// accidentally-stringified env value (e.g. `HOME=undefined` from a
+// shell wrapper) doesn't end up as a literal directory component when
+// the user authored `~/secret`. Combined with the user.Current()
+// fallback further down (see TestOpenClawHome_FallsBackToUserDatabase),
+// the contract is: a malformed HOME falls through to USERPROFILE first,
+// and only if that's also unset/sentinel do we go to the user database.
+func TestOpenClawHome_OSHomeNormalization(t *testing.T) {
+	isolateRuntimeWrites(t)
+	userProfileDir := t.TempDir()
+	homeWinsDir := t.TempDir()
+
+	tests := []struct {
+		name        string
+		home        string
+		userProfile string
+		want        string
+	}{
+		{"HOME=undefined falls through to USERPROFILE", "undefined", userProfileDir, userProfileDir},
+		{"HOME=null falls through to USERPROFILE", "null", userProfileDir, userProfileDir},
+		{"HOME=whitespace falls through to USERPROFILE", "   ", userProfileDir, userProfileDir},
+		{"HOME wins over USERPROFILE when both are valid", homeWinsDir, userProfileDir, homeWinsDir},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			t.Setenv("HOME", tc.home)
+			t.Setenv("USERPROFILE", tc.userProfile)
+			t.Setenv("OPENCLAW_HOME", "")
+			if got := openClawHome(); got != tc.want {
+				t.Errorf("openClawHome() = %q, want %q", got, tc.want)
+			}
+		})
+	}
+}
+
+// TestOpenClawHome_SentinelHOMEFallsToUserDatabaseNotCwd pins the
+// deliberate hybrid documented on osHome: with HOME a sentinel literal
+// and USERPROFILE unset, OpenClaw would fall back to process.cwd();
+// this implementation falls to the OS user database instead. The
+// account home is both safer (cwd-independent) and more useful (it is
+// where the user originally authored `~/...` against), so we prefer it
+// over either OpenClaw's cwd fallback or a strict reject.
+func TestOpenClawHome_SentinelHOMEFallsToUserDatabaseNotCwd(t *testing.T) {
+	isolateRuntimeWrites(t)
+	u, err := user.Current()
+	if err != nil || u.HomeDir == "" {
+		t.Skip("os/user.Current() unavailable on this runner")
+	}
+	t.Setenv("HOME", "undefined")
+	t.Setenv("USERPROFILE", "")
+	t.Setenv("OPENCLAW_HOME", "")
+	got := openClawHome()
+	if got != u.HomeDir {
+		t.Errorf("openClawHome() = %q, want %q (account home, not cwd)", got, u.HomeDir)
+	}
+}
+
+// TestExpandTildePath_BackslashPreservedOnPOSIX pins that `~\secret.json`
+// expands by replacing only the `~` byte, leaving the backslash literally
+// as part of the filename — matching OpenClaw's regex-replace semantics
+// (`/^~(?=$|[\\/])/`) rather than going through filepath.Join (which would
+// drop the backslash on POSIX). On Windows backslash is a real separator,
+// so the literal-byte invariant doesn't apply.
+func TestExpandTildePath_BackslashPreservedOnPOSIX(t *testing.T) {
+	if runtime.GOOS == "windows" {
+		t.Skip("backslash is a path separator on Windows; invariant only applies on POSIX")
+	}
+	fakeHome := t.TempDir()
+	setFakeOSHome(t, fakeHome)
+	t.Setenv("OPENCLAW_HOME", "")
+
+	got := expandTildePath(`~\secret.json`)
+	want := fakeHome + `\secret.json`
+	if got != want {
+		t.Errorf("expandTildePath(%q) = %q, want %q (backslash should be preserved as filename byte)", `~\secret.json`, got, want)
+	}
+}

internal/skillscheck/check.go

@@ -4,9 +4,9 @@
 package skillscheck
 
 // Init runs the synchronous skills version check. Stores a StaleNotice
-// when the local stamp does not match currentVersion. Safe to call
-// from cmd/root.go before rootCmd.Execute(); zero network, zero
-// subprocess — only a local stamp file read.
+// when the local stamp records a version that does not match
+// currentVersion. Safe to call from cmd/root.go before rootCmd.Execute();
+// zero network, zero subprocess — only a local stamp file read.
 //
 // Skip rules: see shouldSkip (CI envs, DEV builds, non-release semver,
 // LARKSUITE_CLI_NO_SKILLS_NOTIFIER opt-out).
@@ -15,10 +15,12 @@ package skillscheck
 //   - shouldSkip rule met
 //   - ReadStamp returns an I/O error other than ENOENT
 //   - Stamp matches currentVersion (in-sync)
+//   - Stamp is missing (cold start) — only users who ran `lark-cli update`
+//     opt into drift tracking; npx-only installs are intentionally silent.
 func Init(currentVersion string) {
 	// Clear any stale notice from a prior call so early returns below
-	// (skip rules / read errors / in-sync) leave pending == nil instead
-	// of preserving a stale value from a previous Init invocation.
+	// (skip rules / read errors / cold start / in-sync) leave pending == nil
+	// instead of preserving a stale value from a previous Init invocation.
 	SetPending(nil)
 	if shouldSkip(currentVersion) {
 		return
@@ -28,11 +30,19 @@ func Init(currentVersion string) {
 		// Fail closed — don't nag for a transient FS problem.
 		return
 	}
+	if stamp == "" {
+		// Cold start: the stamp is written exclusively by `lark-cli update`
+		// (runSkillsAndStamp). Users who installed skills via
+		// `npx skills add larksuite/cli -g` have no stamp yet — they must
+		// not be nagged with "skills not installed", since the on-disk
+		// skills directory may already be fully populated.
+		return
+	}
 	if stamp == currentVersion {
 		return
 	}
 	SetPending(&StaleNotice{
-		Current: stamp, // "" when never synced
+		Current: stamp, // guaranteed non-empty under the new contract
 		Target:  currentVersion,
 	})
 }

internal/skillscheck/check_test.go

@@ -29,17 +29,13 @@ func TestInit_InSync_NoNotice(t *testing.T) {
 	}
 }
 
-func TestInit_ColdStart_NoticeWithEmptyCurrent(t *testing.T) {
+func TestInit_ColdStart_NoNotice(t *testing.T) {
 	clearSkillsSkipEnv(t)
 	resetPending(t)
 	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
 	Init("1.0.21")
-	got := GetPending()
-	if got == nil {
-		t.Fatal("GetPending() = nil, want non-nil for cold start")
-	}
-	if got.Current != "" || got.Target != "1.0.21" {
-		t.Errorf("notice = %+v, want {Current:\"\", Target:\"1.0.21\"}", got)
+	if got := GetPending(); got != nil {
+		t.Errorf("GetPending() = %+v, want nil (cold start is silent)", got)
 	}
 }
 

internal/skillscheck/notice.go

@@ -15,20 +15,20 @@ import (
 
 // StaleNotice signals that the locally synced skills version does not
 // match the running binary. Current is the last successfully synced
-// version (or "" when never synced); Target is the running binary
-// version. Mirrors internal/update.UpdateInfo's pending-notice pattern.
+// version (always non-empty — Init no longer emits a notice on cold
+// start). Target is the running binary version. Mirrors
+// internal/update.UpdateInfo's pending-notice pattern.
 type StaleNotice struct {
 	Current string `json:"current"`
 	Target  string `json:"target"`
 }
 
 // Message returns a single-line, AI-agent-parseable description of the
-// gap plus the canonical fix command. Mirrors internal/update.UpdateInfo.Message
-// in style ("..., run: lark-cli update" suffix).
+// drift plus the canonical fix command. Mirrors internal/update.UpdateInfo.Message
+// in style ("..., run: lark-cli update" suffix). Current is guaranteed
+// non-empty because Init only emits a StaleNotice for the drift case
+// (stamp present and != binary version).
 func (s *StaleNotice) Message() string {
-	if s.Current == "" {
-		return "lark-cli skills not installed, run: lark-cli update"
-	}
 	return fmt.Sprintf(
 		"lark-cli skills %s out of sync with binary %s, run: lark-cli update",
 		s.Current, s.Target,

internal/skillscheck/notice_test.go

@@ -14,11 +14,6 @@ func TestStaleNotice_Message(t *testing.T) {
 		n    StaleNotice
 		want string
 	}{
-		{
-			"cold_start",
-			StaleNotice{Current: "", Target: "1.0.21"},
-			"lark-cli skills not installed, run: lark-cli update",
-		},
 		{
 			"drift",
 			StaleNotice{Current: "1.0.20", Target: "1.0.21"},

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@larksuite/cli",
-  "version": "1.0.28",
+  "version": "1.0.29",
   "description": "The official CLI for Lark/Feishu open platform",
   "bin": {
     "lark-cli": "scripts/run.js"

shortcuts/mail/flag_suggest.go

@@ -0,0 +1,288 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package mail
+
+import (
+	"fmt"
+	"sort"
+	"strings"
+
+	"github.com/spf13/cobra"
+	"github.com/spf13/pflag"
+
+	"github.com/larksuite/cli/internal/output"
+)
+
+// flagName is a package-private snapshot of a pflag.Flag's identity.
+type flagName struct {
+	long, short string
+	hidden      bool
+}
+
+// Candidate is a single suggested flag returned to the user when an
+// unknown flag is detected. It is serialised into the ErrorEnvelope's
+// error.detail.candidates[] array.
+type Candidate struct {
+	// Flag is the long-form spelling of the suggested flag, e.g. "--to".
+	Flag string `json:"flag"`
+	// Shorthand is the single-character shorthand (without the leading
+	// dash) when the suggested flag has one; empty otherwise.
+	Shorthand string `json:"shorthand,omitempty"`
+	// Distance is the Levenshtein edit distance to the unknown token.
+	// Zero indicates a bidirectional prefix hit (Reason == "prefix").
+	Distance int `json:"distance"`
+	// Reason explains how the candidate was matched: "prefix" for
+	// bidirectional prefix hits, "edit_distance" for fuzzy matches.
+	Reason string `json:"reason"`
+}
+
+// maxCandidates caps the number of suggestions returned per error so
+// the JSON envelope stays compact and the user-visible hint remains
+// scannable.
+const maxCandidates = 5
+
+// InstallOnMail attaches the unknown-flag fuzzy-match hook on the mail
+// service cobra parent command. It is invoked exactly once from
+// shortcuts/register.go inside the `service == "mail"` branch.
+//
+// Cobra's FlagErrorFunc walks up the parent chain looking for the nearest
+// non-nil hook, so every mail subcommand inherits this behaviour without
+// any per-shortcut wiring.
+func InstallOnMail(svc *cobra.Command) {
+	if svc == nil {
+		return
+	}
+	svc.SetFlagErrorFunc(flagSuggestErrorFunc)
+}
+
+// flagSuggestErrorFunc converts pflag's unknown-flag errors into a
+// structured *output.ExitError carrying candidate suggestions. Any other
+// error is passed through unchanged so cobra's existing handling kicks in.
+func flagSuggestErrorFunc(c *cobra.Command, err error) error {
+	if err == nil {
+		return nil
+	}
+	token, isShorthand, ok := parseUnknownToken(err.Error())
+	if !ok {
+		// Non unknown-flag errors (e.g. "required flag(s) ... not set")
+		// pass through to cmd/root.go::handleRootError's fallback path.
+		return err
+	}
+	names := collectFlags(c)
+	var matches []Candidate
+	if isShorthand {
+		matches = suggestShorthand(token, names)
+	} else {
+		matches = suggest(token, names)
+	}
+	// Normalise to a non-nil slice so the JSON envelope always emits
+	// `candidates: []` instead of `null`, keeping the wire shape stable
+	// for downstream parsers regardless of command-state.
+	if matches == nil {
+		matches = []Candidate{}
+	}
+	hint := buildHint(c, matches)
+	detail := map[string]any{
+		"unknown":      rawUnknownToken(token, isShorthand),
+		"command_path": c.CommandPath(),
+		"candidates":   matches,
+	}
+	// Code is ExitAPI (=1), matching cobra's default unknown-flag exit
+	// code. The structured type discrimination lives in error.type.
+	return &output.ExitError{
+		Code: output.ExitAPI,
+		Detail: &output.ErrDetail{
+			Type:    "unknown_flag",
+			Message: err.Error(),
+			Hint:    hint,
+			Detail:  detail,
+		},
+	}
+}
+
+// parseUnknownToken extracts the offending flag name from a pflag error
+// string. Recognised forms:
+//
+//   - "unknown flag: --tos"
+//   - "unknown flag: --bogus=val"
+//   - "unknown shorthand flag: 'X' in -Xyz"
+//
+// Anything else returns (_, _, false) so the caller can pass the error
+// through unchanged.
+func parseUnknownToken(errMsg string) (token string, isShorthand bool, ok bool) {
+	const longPrefix = "unknown flag: --"
+	const shortPrefix = "unknown shorthand flag: '"
+	switch {
+	case strings.HasPrefix(errMsg, longPrefix):
+		rest := errMsg[len(longPrefix):]
+		if eq := strings.IndexAny(rest, "= \t"); eq >= 0 {
+			rest = rest[:eq]
+		}
+		return rest, false, rest != ""
+	case strings.HasPrefix(errMsg, shortPrefix):
+		rest := errMsg[len(shortPrefix):]
+		end := strings.IndexByte(rest, '\'')
+		if end <= 0 {
+			return "", false, false
+		}
+		return rest[:end], true, true
+	}
+	return "", false, false
+}
+
+// rawUnknownToken re-attaches the leading dash(es) to a bare token so the
+// JSON envelope echoes the user-visible spelling.
+func rawUnknownToken(token string, isShorthand bool) string {
+	if isShorthand {
+		return "-" + token
+	}
+	return "--" + token
+}
+
+// collectFlags snapshots the merged local + persistent + inherited flag
+// set of cmd. The hidden bit is preserved on each entry; the suggest
+// helpers apply the actual filter so the slice stays reusable.
+func collectFlags(cmd *cobra.Command) []flagName {
+	if cmd == nil {
+		return nil
+	}
+	var out []flagName
+	cmd.Flags().VisitAll(func(f *pflag.Flag) {
+		out = append(out, flagName{long: f.Name, short: f.Shorthand, hidden: f.Hidden})
+	})
+	return out
+}
+
+// suggest produces top-N long-flag candidates for an unknown token, using
+// bidirectional prefix matching first and Levenshtein distance for the
+// remainder. Hidden flags and empty long names are skipped. Results are
+// stably sorted by (Distance asc, Flag asc) and capped at maxCandidates.
+func suggest(unknown string, names []flagName) []Candidate {
+	if unknown == "" || len(names) == 0 {
+		return nil
+	}
+	threshold := levThreshold(unknown)
+	out := make([]Candidate, 0, len(names))
+	seen := make(map[string]struct{}, len(names))
+
+	// Priority 1: bidirectional prefix match.
+	for _, n := range names {
+		if n.hidden || n.long == "" {
+			continue
+		}
+		if strings.HasPrefix(n.long, unknown) || strings.HasPrefix(unknown, n.long) {
+			out = append(out, Candidate{Flag: "--" + n.long, Shorthand: n.short, Distance: 0, Reason: "prefix"})
+			seen[n.long] = struct{}{}
+		}
+	}
+	// Priority 2: Levenshtein distance, skipping already-matched names.
+	for _, n := range names {
+		if n.hidden || n.long == "" {
+			continue
+		}
+		if _, ok := seen[n.long]; ok {
+			continue
+		}
+		if d := levenshtein(unknown, n.long); d <= threshold {
+			out = append(out, Candidate{Flag: "--" + n.long, Shorthand: n.short, Distance: d, Reason: "edit_distance"})
+		}
+	}
+	sort.SliceStable(out, func(i, j int) bool {
+		if out[i].Distance != out[j].Distance {
+			return out[i].Distance < out[j].Distance
+		}
+		return out[i].Flag < out[j].Flag
+	})
+	if len(out) > maxCandidates {
+		out = out[:maxCandidates]
+	}
+	return out
+}
+
+// suggestShorthand produces candidates for an unknown single-character
+// shorthand. It first looks for exact f.Shorthand matches; if there are
+// none, it falls back to long names that begin with the same character.
+// Levenshtein is deliberately not used here since single-char edit
+// distance would match almost every flag.
+func suggestShorthand(c string, names []flagName) []Candidate {
+	if c == "" || len(names) == 0 {
+		return nil
+	}
+	out := make([]Candidate, 0)
+	for _, n := range names {
+		if n.hidden {
+			continue
+		}
+		if n.short == c {
+			out = append(out, Candidate{Flag: "--" + n.long, Shorthand: n.short, Distance: 0, Reason: "prefix"})
+		}
+	}
+	if len(out) == 0 {
+		for _, n := range names {
+			if n.hidden || n.long == "" {
+				continue
+			}
+			if strings.HasPrefix(n.long, c) {
+				out = append(out, Candidate{Flag: "--" + n.long, Shorthand: n.short, Distance: 0, Reason: "prefix"})
+			}
+		}
+	}
+	sort.SliceStable(out, func(i, j int) bool { return out[i].Flag < out[j].Flag })
+	if len(out) > maxCandidates {
+		out = out[:maxCandidates]
+	}
+	return out
+}
+
+// buildHint returns a one-line hint suitable for the ErrorEnvelope.
+// When at least one candidate exists, the top hit is named; otherwise
+// the user is directed to --help.
+func buildHint(c *cobra.Command, matches []Candidate) string {
+	if len(matches) == 0 {
+		return fmt.Sprintf("Run `%s --help` to view available flags", c.CommandPath())
+	}
+	return fmt.Sprintf("Did you mean: %s ?", matches[0].Flag)
+}
+
+// levThreshold returns the maximum acceptable Levenshtein distance for a
+// token of the given length, clamped to [1, 4].
+func levThreshold(s string) int {
+	t := len(s)/3 + 1
+	if t < 1 {
+		return 1
+	}
+	if t > 4 {
+		return 4
+	}
+	return t
+}
+
+// levenshtein computes the standard Levenshtein edit distance between
+// two ASCII strings using a 2-row dynamic-programming table.
+func levenshtein(a, b string) int {
+	la, lb := len(a), len(b)
+	if la == 0 {
+		return lb
+	}
+	if lb == 0 {
+		return la
+	}
+	prev := make([]int, lb+1)
+	curr := make([]int, lb+1)
+	for j := 0; j <= lb; j++ {
+		prev[j] = j
+	}
+	for i := 1; i <= la; i++ {
+		curr[0] = i
+		for j := 1; j <= lb; j++ {
+			cost := 1
+			if a[i-1] == b[j-1] {
+				cost = 0
+			}
+			curr[j] = min(curr[j-1]+1, prev[j]+1, prev[j-1]+cost)
+		}
+		prev, curr = curr, prev
+	}
+	return prev[lb]
+}

shortcuts/mail/flag_suggest_test.go

@@ -0,0 +1,352 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package mail
+
+import (
+	"errors"
+	"testing"
+
+	"github.com/spf13/cobra"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"github.com/larksuite/cli/internal/output"
+)
+
+// --- suggest (long-flag) ---
+
+func TestSuggest_Prefix(t *testing.T) {
+	names := []flagName{
+		{long: "to", short: "t"},
+		{long: "cc"},
+		{long: "subject", short: "s"},
+	}
+	got := suggest("tos", names)
+	require.NotEmpty(t, got)
+	// "tos" has --to as a prefix → bidirectional prefix hit, Distance=0.
+	assert.Equal(t, "--to", got[0].Flag)
+	assert.Equal(t, 0, got[0].Distance)
+	assert.Equal(t, "prefix", got[0].Reason)
+}
+
+func TestSuggest_Levenshtein(t *testing.T) {
+	names := []flagName{
+		{long: "subject"},
+		{long: "body"},
+		{long: "to"},
+	}
+	// Distance 1 from "subject".
+	got := suggest("subjec", names)
+	require.NotEmpty(t, got)
+	// "subjec" is prefix of "subject" → bidirectional prefix.
+	assert.Equal(t, "--subject", got[0].Flag)
+	assert.Equal(t, "prefix", got[0].Reason)
+
+	// True edit-distance: "subjeect" is not a prefix either way of "subject".
+	got = suggest("subjeect", names)
+	require.NotEmpty(t, got)
+	assert.Equal(t, "--subject", got[0].Flag)
+	assert.Equal(t, "edit_distance", got[0].Reason)
+	assert.GreaterOrEqual(t, got[0].Distance, 1)
+}
+
+func TestSuggest_HiddenSkipped(t *testing.T) {
+	names := []flagName{
+		{long: "internal-debug", hidden: true},
+		{long: "interactive"},
+	}
+	got := suggest("internal", names)
+	for _, c := range got {
+		assert.NotEqual(t, "--internal-debug", c.Flag, "hidden flag must not appear in suggestions")
+	}
+}
+
+func TestSuggest_TopNAndStableSort(t *testing.T) {
+	// 6 names all within threshold and at the same distance (1) from the
+	// unknown token so that the lexicographic tiebreak and maxCandidates
+	// cap are both exercised. (Earlier the names were 3-distance from
+	// "zzz" which is above the threshold of 2 — suggest returned empty
+	// and the assertions trivially passed.)
+	names := []flagName{
+		{long: "aaab"},
+		{long: "aaac"},
+		{long: "aaad"},
+		{long: "aaae"},
+		{long: "aaaf"},
+		{long: "aaag"},
+	}
+	got := suggest("aaaa", names)
+	require.Len(t, got, maxCandidates, "must cap at maxCandidates")
+	// All distances equal → lex ordering by Flag asc, top 5 alphabetically.
+	wantFlags := []string{"--aaab", "--aaac", "--aaad", "--aaae", "--aaaf"}
+	gotFlags := []string{got[0].Flag, got[1].Flag, got[2].Flag, got[3].Flag, got[4].Flag}
+	assert.Equal(t, wantFlags, gotFlags, "tiebreak must order by Flag asc")
+}
+
+// --- suggestShorthand ---
+
+func TestSuggestShorthand_Exact(t *testing.T) {
+	names := []flagName{
+		{long: "to", short: "t"},
+		{long: "cc", short: "c"},
+		{long: "subject", short: "s"},
+	}
+	got := suggestShorthand("t", names)
+	require.NotEmpty(t, got)
+	assert.Equal(t, "--to", got[0].Flag)
+	assert.Equal(t, "t", got[0].Shorthand)
+	assert.Equal(t, "prefix", got[0].Reason)
+}
+
+func TestSuggestShorthand_PrefixFallback(t *testing.T) {
+	// No short matches "x"; fall back to long names starting with "x".
+	names := []flagName{
+		{long: "xargs"},
+		{long: "xterm"},
+		{long: "yargs"},
+	}
+	got := suggestShorthand("x", names)
+	require.NotEmpty(t, got)
+	flags := make([]string, 0, len(got))
+	for _, c := range got {
+		flags = append(flags, c.Flag)
+	}
+	assert.Contains(t, flags, "--xargs")
+	assert.Contains(t, flags, "--xterm")
+	assert.NotContains(t, flags, "--yargs")
+}
+
+// --- parseUnknownToken ---
+
+func TestParseUnknownToken_Long(t *testing.T) {
+	tok, isShort, ok := parseUnknownToken("unknown flag: --tos")
+	assert.True(t, ok)
+	assert.False(t, isShort)
+	assert.Equal(t, "tos", tok)
+
+	tok, isShort, ok = parseUnknownToken("unknown flag: --bogus=val")
+	assert.True(t, ok)
+	assert.False(t, isShort)
+	assert.Equal(t, "bogus", tok, "must strip =value tail")
+
+	tok, _, ok = parseUnknownToken("unknown flag: --bogus value")
+	assert.True(t, ok)
+	assert.Equal(t, "bogus", tok, "must strip whitespace tail")
+}
+
+func TestParseUnknownToken_Shorthand(t *testing.T) {
+	tok, isShort, ok := parseUnknownToken("unknown shorthand flag: 'X' in -X")
+	assert.True(t, ok)
+	assert.True(t, isShort)
+	assert.Equal(t, "X", tok)
+
+	tok, isShort, ok = parseUnknownToken("unknown shorthand flag: 'q' in -qrs")
+	assert.True(t, ok)
+	assert.True(t, isShort)
+	assert.Equal(t, "q", tok)
+}
+
+func TestParseUnknownToken_NotMatch(t *testing.T) {
+	cases := []string{
+		`required flag(s) "to" not set`,
+		"some unrelated error",
+		"",
+		"unknown command \"foo\" for \"mail\"",
+	}
+	for _, in := range cases {
+		tok, isShort, ok := parseUnknownToken(in)
+		assert.False(t, ok, "input %q must not match", in)
+		assert.False(t, isShort)
+		assert.Equal(t, "", tok)
+	}
+}
+
+// --- flagSuggestErrorFunc ---
+
+// newFakeMailCmd builds a cobra command tree resembling the mail parent
+// with a handful of flags exercised by the hook tests.
+func newFakeMailCmd() *cobra.Command {
+	c := &cobra.Command{Use: "mail"}
+	c.Flags().String("to", "", "recipients")
+	c.Flags().String("cc", "", "cc recipients")
+	c.Flags().String("subject", "", "subject")
+	c.Flags().StringP("body", "b", "", "body")
+	return c
+}
+
+func TestFlagSuggestErrorFunc_LongUnknown_ReturnsExitError(t *testing.T) {
+	cmd := newFakeMailCmd()
+	got := flagSuggestErrorFunc(cmd, errors.New("unknown flag: --tos"))
+
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr), "expected *output.ExitError, got %T", got)
+	require.NotNil(t, exitErr.Detail)
+	assert.Equal(t, "unknown_flag", exitErr.Detail.Type)
+	assert.Equal(t, "unknown flag: --tos", exitErr.Detail.Message)
+	assert.Contains(t, exitErr.Detail.Hint, "--to")
+
+	detail, ok := exitErr.Detail.Detail.(map[string]any)
+	require.True(t, ok, "Detail.Detail should be map[string]any")
+	assert.Equal(t, "--tos", detail["unknown"])
+	assert.Equal(t, cmd.CommandPath(), detail["command_path"])
+
+	cands, ok := detail["candidates"].([]Candidate)
+	require.True(t, ok, "candidates should be []Candidate")
+	require.NotEmpty(t, cands)
+
+	var foundTo bool
+	for _, c := range cands {
+		if c.Flag == "--to" {
+			foundTo = true
+			assert.Equal(t, "prefix", c.Reason)
+			break
+		}
+	}
+	assert.True(t, foundTo, "expected --to in candidates")
+}
+
+func TestFlagSuggestErrorFunc_NotUnknownFlag_PassesThrough(t *testing.T) {
+	cmd := newFakeMailCmd()
+	in := errors.New(`required flag(s) "to" not set`)
+	got := flagSuggestErrorFunc(cmd, in)
+	// Identity passthrough: same error pointer.
+	assert.Same(t, in, got, "non-unknown-flag errors must be returned unchanged")
+}
+
+func TestFlagSuggestErrorFunc_ExitCodeIsOne(t *testing.T) {
+	cmd := newFakeMailCmd()
+	got := flagSuggestErrorFunc(cmd, errors.New("unknown flag: --tos"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr))
+	// Hard contract — both compile-time and runtime guards:
+	assert.Equal(t, output.ExitAPI, exitErr.Code, "unknown_flag must use ExitAPI, not ExitValidation")
+	assert.Equal(t, 1, output.ExitAPI, "ExitAPI constant must remain 1")
+}
+
+// --- edge-case coverage ---
+
+func TestInstallOnMail_NilIsNoop(t *testing.T) {
+	// Must not panic; the nil-guard is the contract.
+	InstallOnMail(nil)
+}
+
+func TestInstallOnMail_InstallsHook(t *testing.T) {
+	c := newFakeMailCmd()
+	InstallOnMail(c)
+	require.NotNil(t, c.FlagErrorFunc())
+	got := c.FlagErrorFunc()(c, errors.New("unknown flag: --tos"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr), "installed hook must produce *output.ExitError")
+	assert.Equal(t, "unknown_flag", exitErr.Detail.Type)
+}
+
+func TestFlagSuggestErrorFunc_NilError(t *testing.T) {
+	cmd := newFakeMailCmd()
+	assert.NoError(t, flagSuggestErrorFunc(cmd, nil))
+}
+
+func TestFlagSuggestErrorFunc_LongUnknown_StripsValueTail(t *testing.T) {
+	cmd := newFakeMailCmd()
+	got := flagSuggestErrorFunc(cmd, errors.New("unknown flag: --tos=alice@example.com"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr))
+	detail := exitErr.Detail.Detail.(map[string]any)
+	assert.Equal(t, "--tos", detail["unknown"], "value tail must be stripped before echoing")
+}
+
+func TestFlagSuggestErrorFunc_ShorthandUnknown(t *testing.T) {
+	cmd := newFakeMailCmd()
+	got := flagSuggestErrorFunc(cmd, errors.New("unknown shorthand flag: 'b' in -bXY"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr))
+	detail := exitErr.Detail.Detail.(map[string]any)
+	assert.Equal(t, "-b", detail["unknown"])
+	cands, ok := detail["candidates"].([]Candidate)
+	require.True(t, ok)
+	// newFakeMailCmd has --body/-b; exact shorthand hit expected.
+	require.NotEmpty(t, cands)
+	assert.Equal(t, "--body", cands[0].Flag)
+	assert.Equal(t, "b", cands[0].Shorthand)
+}
+
+func TestFlagSuggestErrorFunc_CandidatesAlwaysArray(t *testing.T) {
+	// A cobra command with no flags forces collectFlags → empty names →
+	// suggest → nil. The envelope must still expose candidates as a
+	// non-nil []Candidate so the JSON wire shape is "candidates: []"
+	// rather than "candidates: null".
+	bare := &cobra.Command{Use: "mail"}
+	got := flagSuggestErrorFunc(bare, errors.New("unknown flag: --bogus"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr))
+	detail := exitErr.Detail.Detail.(map[string]any)
+	cands, ok := detail["candidates"].([]Candidate)
+	require.True(t, ok, "candidates must be []Candidate even when empty")
+	assert.NotNil(t, cands, "candidates must be non-nil empty slice, not nil")
+	assert.Empty(t, cands)
+}
+
+func TestFlagSuggestErrorFunc_NoCandidatesUsesHelpHint(t *testing.T) {
+	cmd := newFakeMailCmd()
+	// Token with no plausible neighbor in {to, cc, subject, body}.
+	got := flagSuggestErrorFunc(cmd, errors.New("unknown flag: --zzzzzzz"))
+	var exitErr *output.ExitError
+	require.True(t, errors.As(got, &exitErr))
+	assert.Contains(t, exitErr.Detail.Hint, "--help")
+}
+
+func TestParseUnknownToken_EmptyAndMalformed(t *testing.T) {
+	// Long form with empty token after the prefix.
+	_, _, ok := parseUnknownToken("unknown flag: --")
+	assert.False(t, ok, "empty long token must not match")
+
+	// Shorthand with no closing quote.
+	_, _, ok = parseUnknownToken("unknown shorthand flag: 'q")
+	assert.False(t, ok, "shorthand without closing quote must not match")
+
+	// Shorthand with empty char between quotes.
+	_, _, ok = parseUnknownToken("unknown shorthand flag: '' in -")
+	assert.False(t, ok, "empty shorthand token must not match")
+}
+
+func TestSuggest_EmptyInputs(t *testing.T) {
+	assert.Nil(t, suggest("", []flagName{{long: "to"}}))
+	assert.Nil(t, suggest("foo", nil))
+}
+
+func TestSuggestShorthand_EmptyInputs(t *testing.T) {
+	assert.Nil(t, suggestShorthand("", []flagName{{long: "to", short: "t"}}))
+	assert.Nil(t, suggestShorthand("x", nil))
+}
+
+func TestSuggestShorthand_HiddenSkipped(t *testing.T) {
+	names := []flagName{
+		{long: "secret", short: "s", hidden: true},
+		{long: "subject", short: "s"},
+	}
+	got := suggestShorthand("s", names)
+	for _, c := range got {
+		assert.NotEqual(t, "--secret", c.Flag, "hidden shorthand must not be suggested")
+	}
+}
+
+func TestCollectFlags_NilSafe(t *testing.T) {
+	assert.Nil(t, collectFlags(nil))
+}
+
+func TestLevThreshold_Clamp(t *testing.T) {
+	// len 0 → 0/3+1 = 1
+	assert.Equal(t, 1, levThreshold(""))
+	// len 3 → 2
+	assert.Equal(t, 2, levThreshold("abc"))
+	// Long token caps at 4.
+	assert.Equal(t, 4, levThreshold("aaaaaaaaaaaaaaaaaaaa"))
+}
+
+func TestLevenshtein_EmptyAndIdentical(t *testing.T) {
+	assert.Equal(t, 0, levenshtein("", ""))
+	assert.Equal(t, 3, levenshtein("", "abc"))
+	assert.Equal(t, 3, levenshtein("abc", ""))
+	assert.Equal(t, 0, levenshtein("abc", "abc"))
+	assert.Equal(t, 1, levenshtein("abc", "abd"))
+}

shortcuts/register.go

@@ -99,5 +99,8 @@ func RegisterShortcutsWithContext(ctx context.Context, program *cobra.Command, f
 		for _, shortcut := range shortcuts {
 			shortcut.MountWithContext(ctx, svc, f)
 		}
+		if service == "mail" {
+			mail.InstallOnMail(svc)
+		}
 	}
 }

shortcuts/register_test.go

@@ -6,6 +6,7 @@ package shortcuts
 import (
 	"bytes"
 	"encoding/json"
+	"errors"
 	"fmt"
 	"os"
 	"path/filepath"
@@ -14,6 +15,7 @@ import (
 
 	"github.com/larksuite/cli/internal/cmdutil"
 	"github.com/larksuite/cli/internal/core"
+	"github.com/larksuite/cli/internal/output"
 	"github.com/spf13/cobra"
 )
 
@@ -305,6 +307,65 @@ func TestRegisterShortcutsReusesExistingServiceCommand(t *testing.T) {
 	}
 }
 
+// TestRegisterShortcutsInstallsMailFlagSuggestHook is the end-to-end
+// wiring guard for the mail unknown-flag fuzzy-match feature: it ensures
+// the `if service == "mail" { mail.InstallOnMail(svc) }` branch in
+// RegisterShortcutsWithContext is actually exercised, so a future refactor
+// that drops the branch (or breaks the import) will fail this test rather
+// than silently regressing the structured-error contract.
+func TestRegisterShortcutsInstallsMailFlagSuggestHook(t *testing.T) {
+	program := &cobra.Command{Use: "root"}
+	RegisterShortcuts(program, newRegisterTestFactory(t))
+
+	mailCmd, _, err := program.Find([]string{"mail"})
+	if err != nil {
+		t.Fatalf("find mail command: %v", err)
+	}
+	if mailCmd == nil || mailCmd.Name() != "mail" {
+		t.Fatalf("mail command not mounted: %#v", mailCmd)
+	}
+
+	// The FlagErrorFunc lookup walks up to the nearest non-nil hook, so
+	// invoking it on the mail parent (or any of its children) must yield
+	// a structured *output.ExitError with type "unknown_flag".
+	got := mailCmd.FlagErrorFunc()(mailCmd, errors.New("unknown flag: --bogus"))
+	var exitErr *output.ExitError
+	if !errors.As(got, &exitErr) {
+		t.Fatalf("expected *output.ExitError, got %T (%v)", got, got)
+	}
+	if exitErr.Detail == nil || exitErr.Detail.Type != "unknown_flag" {
+		t.Fatalf("expected Detail.Type=unknown_flag, got %#v", exitErr.Detail)
+	}
+	if exitErr.Code != output.ExitAPI {
+		t.Fatalf("expected Code=ExitAPI(%d), got %d", output.ExitAPI, exitErr.Code)
+	}
+}
+
+// TestRegisterShortcutsLeavesNonMailFlagErrorUntouched confirms the
+// install is scoped: a non-mail service must keep the default cobra
+// pass-through behaviour, otherwise an accidental fall-through in
+// register.go would silently change every domain's error envelope.
+func TestRegisterShortcutsLeavesNonMailFlagErrorUntouched(t *testing.T) {
+	program := &cobra.Command{Use: "root"}
+	RegisterShortcuts(program, newRegisterTestFactory(t))
+
+	baseCmd, _, err := program.Find([]string{"base"})
+	if err != nil {
+		t.Fatalf("find base command: %v", err)
+	}
+	in := errors.New("unknown flag: --bogus")
+	got := baseCmd.FlagErrorFunc()(baseCmd, in)
+	// Default cobra hook is identity — anything else means the mail hook
+	// leaked across domains.
+	var exitErr *output.ExitError
+	if errors.As(got, &exitErr) {
+		t.Fatalf("base service unexpectedly produced *output.ExitError: %#v", exitErr)
+	}
+	if got != in {
+		t.Fatalf("base service should pass through original error pointer, got %T (%v)", got, got)
+	}
+}
+
 func TestGenerateShortcutsJSON(t *testing.T) {
 	output := os.Getenv("SHORTCUTS_OUTPUT")
 	if output == "" {

shortcuts/vc/shortcuts.go

@@ -11,5 +11,8 @@ func Shortcuts() []common.Shortcut {
 		VCSearch,
 		VCNotes,
 		VCRecording,
+		VCMeetingJoin,
+		VCMeetingLeave,
+		VCMeetingEvents,
 	}
 }

shortcuts/vc/vc_meeting_events.go

@@ -0,0 +1,984 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package vc
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"net/http"
+	"sort"
+	"strconv"
+	"strings"
+	"time"
+	"unicode"
+
+	larkcore "github.com/larksuite/oapi-sdk-go/v3/core"
+
+	"github.com/larksuite/cli/internal/output"
+	"github.com/larksuite/cli/shortcuts/common"
+)
+
+const (
+	vcMeetingEventsAPIPath     = "/open-apis/vc/v1/bots/events"
+	defaultVCMeetingEventsSize = 20
+	minVCMeetingEventsPageSize = 20
+	maxVCMeetingEventsPageSize = 100
+	maxVCMeetingEventsPages    = 200
+)
+
+var meetingDisplayLocation = time.FixedZone("UTC+8", 8*60*60)
+
+// toUnixSeconds converts a supported CLI time input into a Unix seconds string.
+func toUnixSeconds(input string, hint ...string) (string, error) {
+	ts, err := common.ParseTime(input, hint...)
+	if err != nil {
+		return "", err
+	}
+	if _, err := strconv.ParseInt(ts, 10, 64); err != nil {
+		return "", fmt.Errorf("invalid timestamp %q: %w", ts, err)
+	}
+	return ts, nil
+}
+
+// VCMeetingEvents lists bot meeting events for a meeting.
+var VCMeetingEvents = common.Shortcut{
+	Service:     "vc",
+	Command:     "+meeting-events",
+	Description: "List bot meeting events by meeting ID",
+	Risk:        "read",
+	Scopes:      []string{"vc:meeting.meetingevent:read"},
+	AuthTypes:   []string{"user"},
+	HasFormat:   true,
+	Flags: []common.Flag{
+		{Name: "meeting-id", Required: true, Desc: "meeting ID to query"},
+		{Name: "start", Desc: "time lower bound (ISO 8601, YYYY-MM-DD, or Unix seconds)"},
+		{Name: "end", Desc: "time upper bound (ISO 8601, YYYY-MM-DD, or Unix seconds)"},
+		{Name: "page-token", Desc: "page token for the next page"},
+		{Name: "page-size", Default: "20", Desc: "page size, 20-100 (default 20)"},
+		{Name: "page-all", Type: "bool", Desc: "automatically paginate through all available pages"},
+	},
+	Validate: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		if err := validateMeetingEventsMeetingID(runtime.Str("meeting-id")); err != nil {
+			return err
+		}
+		if _, err := meetingEventsPageSize(runtime); err != nil {
+			return err
+		}
+		if _, _, err := parseMeetingEventsTimeRange(runtime); err != nil {
+			return err
+		}
+		return nil
+	},
+	DryRun: func(ctx context.Context, runtime *common.RuntimeContext) *common.DryRunAPI {
+		startTime, endTime, err := parseMeetingEventsTimeRange(runtime)
+		if err != nil {
+			return common.NewDryRunAPI().Set("error", err.Error())
+		}
+		params, err := buildMeetingEventsParams(runtime, startTime, endTime)
+		if err != nil {
+			return common.NewDryRunAPI().Set("error", err.Error())
+		}
+		dryRun := common.NewDryRunAPI()
+		if runtime.Bool("page-all") {
+			dryRun = dryRun.Desc("Auto-paginates through all available pages")
+		}
+		dryRun = dryRun.GET(vcMeetingEventsAPIPath)
+		if flat := flattenQueryParams(params); len(flat) > 0 {
+			dryRun.Params(flat)
+		}
+		return dryRun
+	},
+	Execute: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		startTime, endTime, err := parseMeetingEventsTimeRange(runtime)
+		if err != nil {
+			return err
+		}
+		data, events, hasMore, pageToken, err := fetchMeetingEvents(ctx, runtime, startTime, endTime)
+		if err != nil {
+			return err
+		}
+		events = compactMeetingEvents(events)
+		outData := map[string]interface{}{
+			"events":     events,
+			"has_more":   data["has_more"],
+			"page_token": data["page_token"],
+		}
+
+		timeline := buildMeetingEventTimeline(events)
+		runtime.OutFormat(outData, &output.Meta{Count: len(events)}, func(w io.Writer) {
+			if len(timeline.entries) == 0 {
+				fmt.Fprintln(w, "No meeting events.")
+				return
+			}
+			io.WriteString(w, renderMeetingEventsPretty(timeline))
+		})
+		if runtime.Format == "pretty" && pageToken != "" {
+			fmt.Fprintf(runtime.IO().Out, "\npage_token: %s\n", pageToken)
+			if hasMore {
+				fmt.Fprintln(runtime.IO().Out, "more available")
+			}
+		}
+		return nil
+	},
+}
+
+func meetingEventsPageSize(runtime *common.RuntimeContext) (int, error) {
+	if runtime.Bool("page-all") {
+		return maxVCMeetingEventsPageSize, nil
+	}
+	pageSizeStr := strings.TrimSpace(runtime.Str("page-size"))
+	if pageSizeStr == "" {
+		return defaultVCMeetingEventsSize, nil
+	}
+	pageSize, err := strconv.Atoi(pageSizeStr)
+	if err != nil {
+		return 0, common.FlagErrorf("invalid --page-size %q: must be an integer", pageSizeStr)
+	}
+	if pageSize < minVCMeetingEventsPageSize {
+		return minVCMeetingEventsPageSize, nil
+	}
+	if pageSize > maxVCMeetingEventsPageSize {
+		return maxVCMeetingEventsPageSize, nil
+	}
+	return pageSize, nil
+}
+
+func meetingEventsPaginationConfig(runtime *common.RuntimeContext) (bool, int) {
+	if !runtime.Bool("page-all") {
+		return false, 0
+	}
+	return true, maxVCMeetingEventsPages
+}
+
+func validateMeetingEventsMeetingID(meetingID string) error {
+	meetingID = strings.TrimSpace(meetingID)
+	if meetingID == "" {
+		return common.FlagErrorf("--meeting-id is required")
+	}
+	value, err := strconv.ParseInt(meetingID, 10, 64)
+	if err != nil || value <= 0 {
+		return common.FlagErrorf("--meeting-id must be a positive integer, got %q", meetingID)
+	}
+	return nil
+}
+
+// parseMeetingEventsTimeRange validates --start/--end and returns Unix seconds strings.
+func parseMeetingEventsTimeRange(runtime *common.RuntimeContext) (string, string, error) {
+	start := strings.TrimSpace(runtime.Str("start"))
+	end := strings.TrimSpace(runtime.Str("end"))
+	if start == "" && end == "" {
+		return "", "", nil
+	}
+
+	var startTime, endTime string
+	if start != "" {
+		parsed, err := toUnixSeconds(start)
+		if err != nil {
+			return "", "", output.ErrValidation("--start: %v", err)
+		}
+		startTime = parsed
+	}
+	if end != "" {
+		parsed, err := toUnixSeconds(end, "end")
+		if err != nil {
+			return "", "", output.ErrValidation("--end: %v", err)
+		}
+		endTime = parsed
+	}
+	if startTime != "" && endTime != "" {
+		startValue, _ := strconv.ParseInt(startTime, 10, 64)
+		endValue, _ := strconv.ParseInt(endTime, 10, 64)
+		if startValue > endValue {
+			return "", "", output.ErrValidation("--start (%s) is after --end (%s)", start, end)
+		}
+	}
+	return startTime, endTime, nil
+}
+
+func buildMeetingEventsParams(runtime *common.RuntimeContext, startTime, endTime string) (larkcore.QueryParams, error) {
+	pageSize, err := meetingEventsPageSize(runtime)
+	if err != nil {
+		return nil, err
+	}
+
+	params := make(larkcore.QueryParams)
+	params.Set("meeting_id", strings.TrimSpace(runtime.Str("meeting-id")))
+	params.Set("page_size", strconv.Itoa(pageSize))
+	if pageToken := strings.TrimSpace(runtime.Str("page-token")); pageToken != "" {
+		params.Set("page_token", pageToken)
+	}
+	if startTime != "" {
+		params.Set("start_time", startTime)
+	}
+	if endTime != "" {
+		params.Set("end_time", endTime)
+	}
+	return params, nil
+}
+
+func fetchMeetingEvents(ctx context.Context, runtime *common.RuntimeContext, startTime, endTime string) (map[string]interface{}, []interface{}, bool, string, error) {
+	params, err := buildMeetingEventsParams(runtime, startTime, endTime)
+	if err != nil {
+		return nil, nil, false, "", err
+	}
+	autoPaginate, pageLimit := meetingEventsPaginationConfig(runtime)
+	if !autoPaginate {
+		data, err := runtime.DoAPIJSON(http.MethodGet, vcMeetingEventsAPIPath, params, nil)
+		if err != nil {
+			return nil, nil, false, "", err
+		}
+		if data == nil {
+			data = map[string]interface{}{}
+		}
+		events := common.GetSlice(data, "events")
+		hasMore, _ := data["has_more"].(bool)
+		pageToken, _ := data["page_token"].(string)
+		return data, events, hasMore, pageToken, nil
+	}
+
+	var (
+		allEvents     []interface{}
+		lastData      map[string]interface{}
+		lastPageToken string
+		lastHasMore   bool
+	)
+	for page := 0; page < pageLimit; page++ {
+		data, err := runtime.DoAPIJSON(http.MethodGet, vcMeetingEventsAPIPath, params, nil)
+		if err != nil {
+			return nil, nil, false, "", err
+		}
+		if data == nil {
+			data = map[string]interface{}{}
+		}
+		lastData = data
+		events := common.GetSlice(data, "events")
+		allEvents = append(allEvents, events...)
+		lastHasMore, _ = data["has_more"].(bool)
+		lastPageToken, _ = data["page_token"].(string)
+		if !lastHasMore || lastPageToken == "" {
+			break
+		}
+		params.Set("page_token", lastPageToken)
+	}
+	if lastData == nil {
+		lastData = map[string]interface{}{}
+	}
+	lastData["events"] = allEvents
+	lastData["has_more"] = lastHasMore
+	lastData["page_token"] = lastPageToken
+	return lastData, allEvents, lastHasMore, lastPageToken, nil
+}
+
+func flattenQueryParams(params larkcore.QueryParams) map[string]interface{} {
+	if len(params) == 0 {
+		return nil
+	}
+	flat := make(map[string]interface{}, len(params))
+	for key, values := range params {
+		switch len(values) {
+		case 0:
+			continue
+		case 1:
+			flat[key] = values[0]
+		default:
+			copied := make([]string, len(values))
+			copy(copied, values)
+			flat[key] = copied
+		}
+	}
+	return flat
+}
+
+func compactMeetingEvents(events []interface{}) []interface{} {
+	compacted := make([]interface{}, 0, len(events))
+	for _, raw := range events {
+		event, _ := raw.(map[string]interface{})
+		if event == nil {
+			continue
+		}
+		if payload := common.GetMap(event, "payload"); payload != nil {
+			event["payload"] = compactMeetingPayload(payload)
+		}
+		compacted = append(compacted, event)
+	}
+	return compacted
+}
+
+func compactMeetingPayload(payload map[string]interface{}) map[string]interface{} {
+	if payload == nil {
+		return nil
+	}
+	compacted := make(map[string]interface{}, len(payload))
+	for key, value := range payload {
+		if items, ok := value.([]interface{}); ok && len(items) == 0 {
+			continue
+		}
+		compacted[key] = value
+	}
+	return compacted
+}
+
+type meetingTimeline struct {
+	topic     string
+	startTime time.Time
+	hasStart  bool
+	endTime   time.Time
+	hasEnd    bool
+	entries   []meetingTimelineEntry
+}
+
+type meetingTimelineEntry struct {
+	when        time.Time
+	hasWhen     bool
+	sequence    int
+	group       int
+	subject     string
+	description string
+	details     []string
+}
+
+func buildMeetingEventTimeline(events []interface{}) meetingTimeline {
+	timeline := meetingTimeline{}
+	var sequence int
+	var group int
+	for _, raw := range events {
+		event, _ := raw.(map[string]interface{})
+		if event == nil {
+			continue
+		}
+		payload := common.GetMap(event, "payload")
+		if payload == nil {
+			continue
+		}
+		if timeline.topic == "" || !timeline.hasStart || !timeline.hasEnd {
+			populateMeetingHeader(&timeline, common.GetMap(payload, "meeting"))
+		}
+		for _, entry := range buildTimelineEntriesForEvent(event, &sequence, group) {
+			timeline.entries = append(timeline.entries, entry)
+		}
+		group++
+	}
+	sort.SliceStable(timeline.entries, func(i, j int) bool {
+		left := timeline.entries[i]
+		right := timeline.entries[j]
+		switch {
+		case left.hasWhen && right.hasWhen:
+			if left.when.Equal(right.when) {
+				return left.sequence < right.sequence
+			}
+			return left.when.Before(right.when)
+		case left.hasWhen:
+			return true
+		case right.hasWhen:
+			return false
+		default:
+			return left.sequence < right.sequence
+		}
+	})
+	return timeline
+}
+
+func populateMeetingHeader(timeline *meetingTimeline, meeting map[string]interface{}) {
+	if timeline == nil || meeting == nil {
+		return
+	}
+	if timeline.topic == "" {
+		timeline.topic = common.GetString(meeting, "topic")
+	}
+	if !timeline.hasStart {
+		if parsed, ok := parseFlexibleTime(common.GetString(meeting, "start_time")); ok {
+			timeline.startTime = parsed
+			timeline.hasStart = true
+		}
+	}
+	if !timeline.hasEnd {
+		if parsed, ok := parseFlexibleTime(common.GetString(meeting, "end_time")); ok {
+			timeline.endTime = parsed
+			timeline.hasEnd = true
+		}
+	}
+}
+
+func buildTimelineEntriesForEvent(event map[string]interface{}, sequence *int, group int) []meetingTimelineEntry {
+	payload := common.GetMap(event, "payload")
+	if payload == nil {
+		return nil
+	}
+	eventType := meetingEventType(event)
+	eventTime, eventTimeOK := parseFlexibleTime(common.GetString(event, "event_time"))
+	switch eventType {
+	case "participant_joined":
+		return participantJoinedEntries(payload, eventTime, eventTimeOK, sequence, group)
+	case "participant_left":
+		return participantLeftEntries(payload, eventTime, eventTimeOK, sequence, group)
+	case "transcript_received":
+		return transcriptEntries(payload, eventTime, eventTimeOK, sequence, group)
+	case "chat_received":
+		return chatEntries(payload, eventTime, eventTimeOK, sequence, group)
+	case "magic_share_started":
+		return magicShareStartedEntries(payload, eventTime, eventTimeOK, sequence, group)
+	case "magic_share_ended":
+		return magicShareEndedEntries(payload, eventTime, eventTimeOK, sequence, group)
+	default:
+		return []meetingTimelineEntry{newTimelineEntry(eventTime, eventTimeOK, sequence, group, meetingEventUserDisplayName(nil), meetingEventSummary(event), nil)}
+	}
+}
+
+func participantJoinedEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "participant_joined_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "加入了会议", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "join_time"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "participant"))
+		if subject == "" {
+			subject = "未知参会人"
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, "加入了会议", nil))
+	}
+	return entries
+}
+
+func participantLeftEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "participant_left_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "离开了会议", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "leave_time"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "participant"))
+		if subject == "" {
+			subject = "未知参会人"
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, leaveAction(item), nil))
+	}
+	return entries
+}
+
+func transcriptEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "transcript_received_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "产生了转写", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "start_time_ms"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "speaker"))
+		if subject == "" {
+			subject = "未知发言人"
+		}
+		text := strings.TrimSpace(common.GetString(item, "text"))
+		description := "产生了转写"
+		if text != "" {
+			description = text
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, description, nil))
+	}
+	return entries
+}
+
+func chatEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "chat_received_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "发送了消息", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "send_time"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "operator"))
+		if subject == "" {
+			subject = "未知发送者"
+		}
+		typeLabel := chatMessageTypeLabel(item)
+		description := strings.TrimSpace(common.GetString(item, "content"))
+		if description == "" {
+			description = fmt.Sprintf("[%s] 发送了消息", typeLabel)
+		} else {
+			description = fmt.Sprintf("[%s] %s", typeLabel, description)
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, description, nil))
+	}
+	return entries
+}
+
+func magicShareStartedEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "magic_share_started_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "开始共享内容", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "time"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "operator"))
+		if subject == "" {
+			subject = "未知用户"
+		}
+		title := strings.TrimSpace(common.GetString(common.GetMap(item, "share_doc"), "title"))
+		url := strings.TrimSpace(common.GetString(common.GetMap(item, "share_doc"), "url"))
+		description := "开始共享内容"
+		if title != "" {
+			description = fmt.Sprintf("开始共享「%s」", title)
+		}
+		var details []string
+		if url != "" {
+			details = append(details, "URL: "+url)
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, description, details))
+	}
+	return entries
+}
+
+func magicShareEndedEntries(payload map[string]interface{}, fallbackTime time.Time, fallbackOK bool, sequence *int, group int) []meetingTimelineEntry {
+	items := common.GetSlice(payload, "magic_share_ended_items")
+	if len(items) == 0 {
+		return []meetingTimelineEntry{newTimelineEntry(fallbackTime, fallbackOK, sequence, group, "", "结束共享", nil)}
+	}
+	entries := make([]meetingTimelineEntry, 0, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		when, ok := parseFlexibleTime(common.GetString(item, "time"))
+		if !ok {
+			when, ok = fallbackTime, fallbackOK
+		}
+		subject := meetingEventUserWithID(common.GetMap(item, "operator"))
+		if subject == "" {
+			subject = "未知用户"
+		}
+		entries = append(entries, newTimelineEntry(when, ok, sequence, group, subject, "结束共享", nil))
+	}
+	return entries
+}
+
+func newTimelineEntry(when time.Time, hasWhen bool, sequence *int, group int, subject, description string, details []string) meetingTimelineEntry {
+	entry := meetingTimelineEntry{
+		when:        when,
+		hasWhen:     hasWhen,
+		sequence:    *sequence,
+		group:       group,
+		subject:     subject,
+		description: description,
+		details:     details,
+	}
+	*sequence = *sequence + 1
+	return entry
+}
+
+func parseFlexibleTime(raw string) (time.Time, bool) {
+	raw = strings.TrimSpace(raw)
+	if raw == "" {
+		return time.Time{}, false
+	}
+	if ts, err := strconv.ParseInt(raw, 10, 64); err == nil {
+		switch {
+		case ts > 1_000_000_000_000:
+			return time.UnixMilli(ts), true
+		case ts > 0:
+			return time.Unix(ts, 0), true
+		}
+	}
+	if parsed, err := time.Parse(time.RFC3339, raw); err == nil {
+		return parsed, true
+	}
+	return time.Time{}, false
+}
+
+func renderMeetingEventsPretty(timeline meetingTimeline) string {
+	var b strings.Builder
+	if timeline.topic != "" {
+		fmt.Fprintf(&b, "会议主题：%s\n", escapePrettyText(timeline.topic))
+	}
+	if timeline.hasStart || timeline.hasEnd {
+		fmt.Fprintf(&b, "会议时间：%s\n", escapePrettyText(formatMeetingWindow(timeline.startTime, timeline.hasStart, timeline.endTime, timeline.hasEnd)))
+	}
+	if b.Len() > 0 {
+		b.WriteString("\n")
+	}
+	for _, entry := range timeline.entries {
+		fmt.Fprintf(&b, "[%s] ", formatTimelineOffset(entry.when, entry.hasWhen, timeline.startTime, timeline.hasStart))
+		if entry.subject != "" {
+			if entry.description == "" {
+				fmt.Fprintln(&b, escapePrettyText(entry.subject))
+				for _, detail := range entry.details {
+					fmt.Fprintf(&b, "           %s\n", escapePrettyText(detail))
+				}
+				continue
+			}
+			if needsColon(entry.description) {
+				fmt.Fprintf(&b, "%s: %s\n", escapePrettyText(entry.subject), escapePrettyText(entry.description))
+			} else {
+				fmt.Fprintf(&b, "%s %s\n", escapePrettyText(entry.subject), escapePrettyText(entry.description))
+			}
+			for _, detail := range entry.details {
+				fmt.Fprintf(&b, "           %s\n", escapePrettyText(detail))
+			}
+			continue
+		}
+		fmt.Fprintln(&b, escapePrettyText(entry.description))
+		for _, detail := range entry.details {
+			fmt.Fprintf(&b, "           %s\n", escapePrettyText(detail))
+		}
+	}
+	if b.Len() == 0 {
+		return ""
+	}
+	return b.String()
+}
+
+func escapePrettyText(s string) string {
+	if s == "" {
+		return ""
+	}
+	var b strings.Builder
+	for _, r := range s {
+		switch r {
+		case '\n':
+			b.WriteString(`\n`)
+		case '\r':
+			b.WriteString(`\r`)
+		case '\t':
+			b.WriteString(`\t`)
+		default:
+			if unicode.IsControl(r) {
+				fmt.Fprintf(&b, "\\u%04X", r)
+				continue
+			}
+			b.WriteRune(r)
+		}
+	}
+	return b.String()
+}
+
+func formatMeetingWindow(start time.Time, hasStart bool, end time.Time, hasEnd bool) string {
+	switch {
+	case hasStart && hasEnd:
+		if !end.After(start) {
+			return fmt.Sprintf("%s（进行中）", start.In(meetingDisplayLocation).Format("2006-01-02 15:04:05"))
+		}
+		return fmt.Sprintf("%s - %s", start.In(meetingDisplayLocation).Format("2006-01-02 15:04:05"), end.In(meetingDisplayLocation).Format("2006-01-02 15:04:05"))
+	case hasStart:
+		return start.In(meetingDisplayLocation).Format("2006-01-02 15:04:05")
+	case hasEnd:
+		return end.In(meetingDisplayLocation).Format("2006-01-02 15:04:05")
+	default:
+		return ""
+	}
+}
+
+func formatTimelineOffset(when time.Time, hasWhen bool, meetingStart time.Time, hasMeetingStart bool) string {
+	if hasWhen && hasMeetingStart {
+		diff := when.Sub(meetingStart)
+		if diff < 0 {
+			diff = 0
+		}
+		totalSeconds := int(diff.Seconds())
+		hours := totalSeconds / 3600
+		minutes := (totalSeconds % 3600) / 60
+		seconds := totalSeconds % 60
+		return fmt.Sprintf("%02d:%02d:%02d", hours, minutes, seconds)
+	}
+	if hasWhen {
+		return when.In(meetingDisplayLocation).Format("15:04:05")
+	}
+	return "??:??:??"
+}
+
+func needsColon(description string) bool {
+	switch description {
+	case "发送了消息", "产生了转写":
+		return false
+	default:
+		return !strings.HasPrefix(description, "加入了") &&
+			!strings.HasPrefix(description, "离开了") &&
+			!strings.HasPrefix(description, "被移出") &&
+			!strings.HasPrefix(description, "会议结束") &&
+			!strings.HasPrefix(description, "开始共享") &&
+			!strings.HasPrefix(description, "结束共享")
+	}
+}
+
+func leaveAction(item map[string]interface{}) string {
+	switch int(common.GetFloat(item, "leave_reason")) {
+	case 2:
+		return "因会议结束离开了会议"
+	case 3:
+		return "被移出了会议"
+	default:
+		return "离开了会议"
+	}
+}
+
+func meetingEventUserWithID(user map[string]interface{}) string {
+	if user == nil {
+		return ""
+	}
+	userID := common.GetString(user, "id")
+	userName := common.GetString(user, "user_name")
+	switch {
+	case userName != "" && userID != "":
+		return fmt.Sprintf("%s(%s)", userName, userID)
+	case userName != "":
+		return userName
+	case userID != "":
+		return userID
+	default:
+		return ""
+	}
+}
+
+func meetingEventType(event map[string]interface{}) string {
+	if eventType := common.GetString(event, "event_type"); eventType != "" {
+		return eventType
+	}
+	return common.GetString(common.GetMap(event, "payload"), "activity_event_type")
+}
+
+func meetingEventSummary(event map[string]interface{}) string {
+	payload := common.GetMap(event, "payload")
+	eventType := meetingEventType(event)
+	switch eventType {
+	case "participant_joined":
+		return participantJoinedSummary(payload)
+	case "participant_left":
+		return participantLeftSummary(payload)
+	case "transcript_received":
+		return transcriptReceivedSummary(payload)
+	case "chat_received":
+		return chatReceivedSummary(payload)
+	case "magic_share_started":
+		return magicShareStartedSummary(payload)
+	case "magic_share_ended":
+		return magicShareEndedSummary(payload)
+	default:
+		return fallbackMeetingEventSummary(payload, eventType)
+	}
+}
+
+func participantJoinedSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "participant_joined_items")
+	switch len(items) {
+	case 0:
+		return "participant joined"
+	case 1:
+		user := common.GetMap(firstSliceMap(payload, "participant_joined_items"), "participant")
+		if label := meetingEventUserLabel(user); label != "" {
+			return fmt.Sprintf("participant %s joined", label)
+		}
+		return "participant joined"
+	default:
+		return fmt.Sprintf("%d participants joined", len(items))
+	}
+}
+
+func participantLeftSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "participant_left_items")
+	switch len(items) {
+	case 0:
+		return "participant left"
+	case 1:
+		user := common.GetMap(firstSliceMap(payload, "participant_left_items"), "participant")
+		if label := meetingEventUserLabel(user); label != "" {
+			return fmt.Sprintf("participant %s left", label)
+		}
+		return "participant left"
+	default:
+		return fmt.Sprintf("%d participants left", len(items))
+	}
+}
+
+func transcriptReceivedSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "transcript_received_items")
+	if len(items) > 1 {
+		return fmt.Sprintf("%d transcript items", len(items))
+	}
+	item := firstSliceMap(payload, "transcript_received_items")
+	text := common.GetString(item, "text")
+	speaker := meetingEventUserLabel(common.GetMap(item, "speaker"))
+	switch {
+	case speaker != "" && text != "":
+		return fmt.Sprintf("speaker %s: %s", speaker, text)
+	case speaker != "":
+		return fmt.Sprintf("speaker %s transcript received", speaker)
+	case text != "":
+		return fmt.Sprintf("transcript: %s", text)
+	default:
+		return "transcript received"
+	}
+}
+
+func chatReceivedSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "chat_received_items")
+	switch len(items) {
+	case 0:
+		return "chat received"
+	case 1:
+		item := firstSliceMap(payload, "chat_received_items")
+		content := common.GetString(item, "content")
+		operator := meetingEventUserDisplayName(common.GetMap(item, "operator"))
+		switch {
+		case operator != "" && content != "":
+			return fmt.Sprintf("%s: %s", operator, content)
+		case operator != "":
+			return fmt.Sprintf("message by %s", operator)
+		case content != "":
+			return fmt.Sprintf("message: %s", content)
+		default:
+			return "chat received"
+		}
+	default:
+		count, operator := summarizeChatOperators(items)
+		switch {
+		case count == 1 && operator != "":
+			return fmt.Sprintf("%d messages by %s", len(items), operator)
+		case count > 1:
+			return fmt.Sprintf("%d messages by %d users", len(items), count)
+		default:
+			return fmt.Sprintf("%d messages", len(items))
+		}
+	}
+}
+
+func magicShareStartedSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "magic_share_started_items")
+	if len(items) > 1 {
+		return fmt.Sprintf("%d share start events", len(items))
+	}
+	item := firstSliceMap(payload, "magic_share_started_items")
+	shareID := common.GetString(item, "share_id")
+	title := common.GetString(common.GetMap(item, "share_doc"), "title")
+	switch {
+	case shareID != "" && title != "":
+		return fmt.Sprintf("share %s started: %s", shareID, title)
+	case shareID != "":
+		return fmt.Sprintf("share %s started", shareID)
+	case title != "":
+		return fmt.Sprintf("share started: %s", title)
+	default:
+		return "share started"
+	}
+}
+
+func magicShareEndedSummary(payload map[string]interface{}) string {
+	items := common.GetSlice(payload, "magic_share_ended_items")
+	if len(items) > 1 {
+		return fmt.Sprintf("%d share end events", len(items))
+	}
+	item := firstSliceMap(payload, "magic_share_ended_items")
+	if shareID := common.GetString(item, "share_id"); shareID != "" {
+		return fmt.Sprintf("share %s ended", shareID)
+	}
+	return "share ended"
+}
+
+func fallbackMeetingEventSummary(payload map[string]interface{}, eventType string) string {
+	meeting := common.GetMap(payload, "meeting")
+	if topic := common.GetString(meeting, "topic"); topic != "" {
+		if eventType != "" {
+			return fmt.Sprintf("%s: %s", eventType, topic)
+		}
+		return topic
+	}
+	if eventType != "" {
+		return eventType
+	}
+	return "meeting event"
+}
+
+func firstSliceMap(payload map[string]interface{}, key string) map[string]interface{} {
+	items := common.GetSlice(payload, key)
+	if len(items) == 0 {
+		return nil
+	}
+	first, _ := items[0].(map[string]interface{})
+	return first
+}
+
+func meetingEventUserLabel(user map[string]interface{}) string {
+	if user == nil {
+		return ""
+	}
+	userID := common.GetString(user, "id")
+	userName := common.GetString(user, "user_name")
+	switch {
+	case userID != "" && userName != "":
+		return fmt.Sprintf("%s (%s)", userID, userName)
+	case userID != "":
+		return userID
+	case userName != "":
+		return userName
+	default:
+		return ""
+	}
+}
+
+func meetingEventUserDisplayName(user map[string]interface{}) string {
+	if user == nil {
+		return ""
+	}
+	if userName := common.GetString(user, "user_name"); userName != "" {
+		return userName
+	}
+	return common.GetString(user, "id")
+}
+
+func chatMessageTypeLabel(item map[string]interface{}) string {
+	code := int(common.GetFloat(item, "message_type"))
+	switch code {
+	case 1:
+		return "text"
+	case 2:
+		return "system"
+	case 3:
+		return "reaction"
+	case 4:
+		return "encrypted"
+	default:
+		return "unknown"
+	}
+}
+
+func summarizeChatOperators(items []interface{}) (int, string) {
+	seen := make(map[string]struct{}, len(items))
+	for _, raw := range items {
+		item, _ := raw.(map[string]interface{})
+		if item == nil {
+			continue
+		}
+		operator := meetingEventUserDisplayName(common.GetMap(item, "operator"))
+		if operator == "" {
+			continue
+		}
+		seen[operator] = struct{}{}
+	}
+	if len(seen) != 1 {
+		return len(seen), ""
+	}
+	for operator := range seen {
+		return 1, operator
+	}
+	return 0, ""
+}

shortcuts/vc/vc_meeting_events_test.go

@@ -0,0 +1,931 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package vc
+
+import (
+	"context"
+	"reflect"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/spf13/cobra"
+
+	"github.com/larksuite/cli/internal/cmdutil"
+	"github.com/larksuite/cli/internal/httpmock"
+	"github.com/larksuite/cli/shortcuts/common"
+	larkcore "github.com/larksuite/oapi-sdk-go/v3/core"
+)
+
+func newMeetingEventsRuntime() *common.RuntimeContext {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-id", "", "")
+	cmd.Flags().String("start", "", "")
+	cmd.Flags().String("end", "", "")
+	cmd.Flags().String("page-token", "", "")
+	cmd.Flags().String("page-size", "", "")
+	cmd.Flags().Bool("page-all", false, "")
+	return common.TestNewRuntimeContext(cmd, defaultConfig())
+}
+
+func mustSetMeetingEventsFlag(t *testing.T, runtime *common.RuntimeContext, name, value string) {
+	t.Helper()
+	if err := runtime.Cmd.Flags().Set(name, value); err != nil {
+		t.Fatalf("Flags().Set(%q, %q) error = %v", name, value, err)
+	}
+}
+
+func meetingEventsStub(events []interface{}, hasMore bool, pageToken string) *httpmock.Stub {
+	return &httpmock.Stub{
+		Method: "GET",
+		URL:    vcMeetingEventsAPIPath,
+		Body: map[string]interface{}{
+			"code": 0,
+			"msg":  "ok",
+			"data": map[string]interface{}{
+				"total":      len(events),
+				"has_more":   hasMore,
+				"page_token": pageToken,
+				"events":     events,
+			},
+		},
+	}
+}
+
+func participantJoinedEvent() map[string]interface{} {
+	return map[string]interface{}{
+		"event_id":   "event-1",
+		"event_type": "participant_joined",
+		"event_time": "2026-04-17T08:00:00Z",
+		"payload": map[string]interface{}{
+			"activity_event_type": "participant_joined",
+			"meeting": map[string]interface{}{
+				"id":         "7628568141510692381",
+				"topic":      "项目例会",
+				"meeting_no": "724939760",
+				"start_time": "1776407700",
+				"end_time":   "1776411300",
+			},
+			"participant_joined_items": []interface{}{
+				map[string]interface{}{
+					"participant": map[string]interface{}{
+						"id":        "bot_001",
+						"user_name": "Demo Bot",
+					},
+					"join_time": "2026-04-17T08:00:00Z",
+				},
+			},
+		},
+	}
+}
+
+func participantJoinedEventOngoing() map[string]interface{} {
+	event := participantJoinedEvent()
+	payload := common.GetMap(event, "payload")
+	meeting := common.GetMap(payload, "meeting")
+	meeting["start_time"] = "1776410100"
+	meeting["end_time"] = "1776410100"
+	return event
+}
+
+func chatReceivedEvent() map[string]interface{} {
+	return map[string]interface{}{
+		"event_id":   "event-2",
+		"event_type": "chat_received",
+		"event_time": "2026-04-17T08:05:00Z",
+		"payload": map[string]interface{}{
+			"activity_event_type": "chat_received",
+			"meeting": map[string]interface{}{
+				"id":         "7628568141510692381",
+				"topic":      "项目例会",
+				"meeting_no": "724939760",
+				"start_time": "1776407700",
+				"end_time":   "1776411300",
+			},
+			"participant_joined_items":  []interface{}{},
+			"participant_left_items":    []interface{}{},
+			"transcript_received_items": []interface{}{},
+			"magic_share_started_items": []interface{}{},
+			"magic_share_ended_items":   []interface{}{},
+			"chat_received_items": []interface{}{
+				map[string]interface{}{
+					"content":      "hello",
+					"message_type": 3,
+					"operator": map[string]interface{}{
+						"id":        "u1",
+						"user_name": "Alice",
+					},
+				},
+			},
+		},
+	}
+}
+
+func multiChatReceivedEvent() map[string]interface{} {
+	return map[string]interface{}{
+		"event_id":   "event-3",
+		"event_type": "chat_received",
+		"event_time": "2026-04-17T08:06:00Z",
+		"payload": map[string]interface{}{
+			"activity_event_type": "chat_received",
+			"meeting": map[string]interface{}{
+				"id":         "7628568141510692381",
+				"topic":      "项目例会",
+				"meeting_no": "724939760",
+				"start_time": "1776407700",
+				"end_time":   "1776411300",
+			},
+			"chat_received_items": []interface{}{
+				map[string]interface{}{
+					"content":      "第一条\n第二行",
+					"message_type": 3,
+					"send_time":    "1776408061000",
+					"operator": map[string]interface{}{
+						"id":        "u1",
+						"user_name": "Alice",
+					},
+				},
+				map[string]interface{}{
+					"content":      "第二条",
+					"message_type": 3,
+					"send_time":    "1776408062000",
+					"operator": map[string]interface{}{
+						"id":        "u1",
+						"user_name": "Alice",
+					},
+				},
+			},
+		},
+	}
+}
+
+func magicShareStartedEvent() map[string]interface{} {
+	return map[string]interface{}{
+		"event_id":   "event-4",
+		"event_type": "magic_share_started",
+		"event_time": "2026-04-17T08:07:00Z",
+		"payload": map[string]interface{}{
+			"activity_event_type": "magic_share_started",
+			"meeting": map[string]interface{}{
+				"id":         "7628568141510692381",
+				"topic":      "项目例会",
+				"meeting_no": "724939760",
+				"start_time": "1776407700",
+				"end_time":   "1776411300",
+			},
+			"magic_share_started_items": []interface{}{
+				map[string]interface{}{
+					"time": "1776408123000",
+					"operator": map[string]interface{}{
+						"id":        "u2",
+						"user_name": "Bob",
+					},
+					"share_doc": map[string]interface{}{
+						"title": "共享文档",
+						"url":   "https://example.com/doc",
+					},
+				},
+			},
+		},
+	}
+}
+
+func TestChatReceivedSummary_MultipleItems(t *testing.T) {
+	payload := map[string]interface{}{
+		"chat_received_items": []interface{}{
+			map[string]interface{}{"content": "hello"},
+			map[string]interface{}{"content": "world"},
+		},
+	}
+
+	got := chatReceivedSummary(payload)
+	if got != "2 messages" {
+		t.Fatalf("chatReceivedSummary() = %q, want %q", got, "2 messages")
+	}
+}
+
+func TestChatReceivedSummary_MultipleItemsSameOperator(t *testing.T) {
+	payload := map[string]interface{}{
+		"chat_received_items": []interface{}{
+			map[string]interface{}{"content": "hello", "operator": map[string]interface{}{"id": "u1", "user_name": "Alice"}},
+			map[string]interface{}{"content": "world", "operator": map[string]interface{}{"id": "u1", "user_name": "Alice"}},
+		},
+	}
+
+	got := chatReceivedSummary(payload)
+	if got != "2 messages by Alice" {
+		t.Fatalf("chatReceivedSummary() = %q, want %q", got, "2 messages by Alice")
+	}
+}
+
+func TestChatReceivedSummary_MultipleItemsMultipleOperators(t *testing.T) {
+	payload := map[string]interface{}{
+		"chat_received_items": []interface{}{
+			map[string]interface{}{"content": "hello", "operator": map[string]interface{}{"id": "u1", "user_name": "Alice"}},
+			map[string]interface{}{"content": "world", "operator": map[string]interface{}{"id": "u2", "user_name": "Bob"}},
+			map[string]interface{}{"content": "again", "operator": map[string]interface{}{"id": "u3", "user_name": "Carol"}},
+		},
+	}
+
+	got := chatReceivedSummary(payload)
+	if got != "3 messages by 3 users" {
+		t.Fatalf("chatReceivedSummary() = %q, want %q", got, "3 messages by 3 users")
+	}
+}
+
+func TestParticipantJoinedSummary_MultipleItems(t *testing.T) {
+	payload := map[string]interface{}{
+		"participant_joined_items": []interface{}{
+			map[string]interface{}{"participant": map[string]interface{}{"id": "u1", "user_name": "User 1"}},
+			map[string]interface{}{"participant": map[string]interface{}{"id": "u2", "user_name": "User 2"}},
+		},
+	}
+
+	got := participantJoinedSummary(payload)
+	if got != "2 participants joined" {
+		t.Fatalf("participantJoinedSummary() = %q, want %q", got, "2 participants joined")
+	}
+}
+
+func TestMeetingEvents_Validation_InvalidMeetingID(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "not-a-number")
+
+	err := VCMeetingEvents.Validate(context.Background(), runtime)
+	if err == nil {
+		t.Fatal("expected validation error for invalid meeting ID")
+	}
+	if !strings.Contains(err.Error(), "positive integer") {
+		t.Fatalf("unexpected error: %v", err)
+	}
+}
+
+func TestMeetingEvents_Validation_InvalidTimeRange(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "start", "200")
+	mustSetMeetingEventsFlag(t, runtime, "end", "100")
+
+	err := VCMeetingEvents.Validate(context.Background(), runtime)
+	if err == nil {
+		t.Fatal("expected validation error for invalid time range")
+	}
+	if !strings.Contains(err.Error(), "after --end") {
+		t.Fatalf("unexpected error: %v", err)
+	}
+}
+
+func TestMeetingEvents_Validation_PageSizeBelowMinDoesNotError(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "10")
+
+	err := VCMeetingEvents.Validate(context.Background(), runtime)
+	if err != nil {
+		t.Fatalf("expected no validation error for page-size clamp, got: %v", err)
+	}
+}
+
+func TestMeetingEvents_Validation_PageAllIgnoresInvalidPageSize(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-all", "true")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "10")
+
+	err := VCMeetingEvents.Validate(context.Background(), runtime)
+	if err != nil {
+		t.Fatalf("expected no validation error when page-all ignores page-size, got: %v", err)
+	}
+}
+
+func TestMeetingEvents_Validation_InvalidPageSizeReturnsFlagError(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "foo")
+
+	err := VCMeetingEvents.Validate(context.Background(), runtime)
+	if err == nil {
+		t.Fatal("expected validation error for non-integer page-size")
+	}
+	if !strings.Contains(err.Error(), "invalid --page-size") {
+		t.Fatalf("unexpected error: %v", err)
+	}
+}
+
+func TestBuildMeetingEventsParams(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "40")
+	mustSetMeetingEventsFlag(t, runtime, "page-token", "1710000000000000000")
+
+	params, err := buildMeetingEventsParams(runtime, "1710000000", "1710003600")
+	if err != nil {
+		t.Fatalf("buildMeetingEventsParams() error = %v", err)
+	}
+	if got := params["meeting_id"][0]; got != "7628568141510692381" {
+		t.Fatalf("meeting_id = %q, want %q", got, "7628568141510692381")
+	}
+	if got := params["page_size"][0]; got != "40" {
+		t.Fatalf("page_size = %q, want %q", got, "40")
+	}
+	if got := params["page_token"][0]; got != "1710000000000000000" {
+		t.Fatalf("page_token = %q, want %q", got, "1710000000000000000")
+	}
+	if got := params["start_time"][0]; got != "1710000000" {
+		t.Fatalf("start_time = %q, want %q", got, "1710000000")
+	}
+	if got := params["end_time"][0]; got != "1710003600" {
+		t.Fatalf("end_time = %q, want %q", got, "1710003600")
+	}
+}
+
+func TestBuildMeetingEventsParams_PageSizeBelowMinClampsToMin(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "10")
+
+	params, err := buildMeetingEventsParams(runtime, "", "")
+	if err != nil {
+		t.Fatalf("buildMeetingEventsParams() error = %v", err)
+	}
+	if got := params["page_size"][0]; got != "20" {
+		t.Fatalf("page_size = %q, want %q when below min", got, "20")
+	}
+}
+
+func TestBuildMeetingEventsParams_PageSizeAboveMaxClampsToMax(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "999")
+
+	params, err := buildMeetingEventsParams(runtime, "", "")
+	if err != nil {
+		t.Fatalf("buildMeetingEventsParams() error = %v", err)
+	}
+	if got := params["page_size"][0]; got != "100" {
+		t.Fatalf("page_size = %q, want %q when above max", got, "100")
+	}
+}
+
+func TestBuildMeetingEventsParams_PageAllUsesMaxPageSize(t *testing.T) {
+	runtime := newMeetingEventsRuntime()
+	mustSetMeetingEventsFlag(t, runtime, "meeting-id", "7628568141510692381")
+	mustSetMeetingEventsFlag(t, runtime, "page-all", "true")
+	mustSetMeetingEventsFlag(t, runtime, "page-size", "50")
+
+	params, err := buildMeetingEventsParams(runtime, "", "")
+	if err != nil {
+		t.Fatalf("buildMeetingEventsParams() error = %v", err)
+	}
+	if got := params["page_size"][0]; got != "100" {
+		t.Fatalf("page_size = %q, want %q when page-all is set", got, "100")
+	}
+}
+
+func TestMeetingEvents_DryRun(t *testing.T) {
+	f, stdout, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--page-token", "1710000000000000000",
+		"--page-size", "40",
+		"--start", "1710000000",
+		"--end", "1710003600",
+		"--dry-run",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	out := stdout.String()
+	for _, want := range []string{
+		vcMeetingEventsAPIPath,
+		`"meeting_id": "7628568141510692381"`,
+		`"page_token": "1710000000000000000"`,
+		`"page_size": "40"`,
+		`"start_time": "1710000000"`,
+		`"end_time": "1710003600"`,
+	} {
+		if !strings.Contains(out, want) {
+			t.Fatalf("dry-run output missing %q: %s", want, out)
+		}
+	}
+}
+
+func TestMeetingEvents_DryRun_PageAllUsesMaxLimit(t *testing.T) {
+	f, stdout, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--page-all",
+		"--dry-run",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	if !strings.Contains(stdout.String(), "Auto-paginates through all available pages") {
+		t.Fatalf("dry-run output missing auto-pagination description: %s", stdout.String())
+	}
+}
+
+func TestMeetingEvents_ExecuteJSON_PageAll(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub([]interface{}{participantJoinedEvent()}, true, "pt_2"))
+	reg.Register(meetingEventsStub([]interface{}{participantJoinedEvent()}, false, ""))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "json",
+		"--page-all",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	out := strings.ReplaceAll(stdout.String(), " ", "")
+	out = strings.ReplaceAll(out, "\n", "")
+	if count := strings.Count(out, `"event_type":"participant_joined"`); count != 2 {
+		t.Fatalf("expected 2 aggregated events, got %d: %s", count, stdout.String())
+	}
+	if !strings.Contains(out, `"has_more":false`) {
+		t.Fatalf("expected final has_more=false: %s", stdout.String())
+	}
+}
+
+func TestMeetingEvents_ExecuteJSON(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub([]interface{}{participantJoinedEvent()}, true, "1710000000000000000"))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "json",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	out := strings.ReplaceAll(stdout.String(), " ", "")
+	out = strings.ReplaceAll(out, "\n", "")
+	for _, want := range []string{
+		`"event_type":"participant_joined"`,
+		`"has_more":true`,
+		`"page_token":"1710000000000000000"`,
+		`"events":[`,
+	} {
+		if !strings.Contains(out, want) {
+			t.Fatalf("json output missing %q: %s", want, stdout.String())
+		}
+	}
+}
+
+func TestMeetingEvents_ExecuteJSON_PrunesEmptySlices(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub([]interface{}{chatReceivedEvent()}, false, ""))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "json",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	out := stdout.String()
+	for _, unwanted := range []string{
+		`"participant_joined_items": []`,
+		`"participant_left_items": []`,
+		`"transcript_received_items": []`,
+		`"magic_share_started_items": []`,
+		`"magic_share_ended_items": []`,
+	} {
+		if strings.Contains(out, unwanted) {
+			t.Fatalf("json output should not contain %q: %s", unwanted, out)
+		}
+	}
+	if !strings.Contains(out, `"message_type": 3`) {
+		t.Fatalf("json output should keep numeric fields: %s", out)
+	}
+}
+
+func TestMeetingEvents_ExecutePretty(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub([]interface{}{participantJoinedEventOngoing(), multiChatReceivedEvent(), magicShareStartedEvent()}, true, "1710000000000000000"))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "pretty",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	out := stdout.String()
+	for _, want := range []string{
+		"会议主题：项目例会",
+		"会议时间：2026-04-17 15:15:00（进行中）",
+		"Demo Bot(bot_001) 加入了会议",
+		"Alice(u1): [reaction] 第一条\\n第二行",
+		"Alice(u1): [reaction] 第二条",
+		"Bob(u2) 开始共享「共享文档」",
+		"URL: https://example.com/doc",
+		"page_token: 1710000000000000000",
+	} {
+		if !strings.Contains(out, want) {
+			t.Fatalf("pretty output missing %q: %s", want, out)
+		}
+	}
+	if strings.Contains(out, "第二条\n\n[") {
+		t.Fatalf("pretty output should not insert blank lines between event entries: %s", out)
+	}
+	if !strings.Contains(out, "第二条\n[") {
+		t.Fatalf("pretty output should keep event entries contiguous: %s", out)
+	}
+}
+
+func TestMeetingEvents_ExecutePretty_PrintsPageTokenWithoutHasMore(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub([]interface{}{participantJoinedEventOngoing()}, false, "pt_last"))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "pretty",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	out := stdout.String()
+	if !strings.Contains(out, "page_token: pt_last") {
+		t.Fatalf("pretty output should print page_token even when has_more is false: %s", out)
+	}
+	if strings.Contains(out, "more available") {
+		t.Fatalf("pretty output should not print more-available hint when has_more is false: %s", out)
+	}
+}
+
+func TestMeetingEvents_ExecuteEmpty(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+	reg.Register(meetingEventsStub(nil, false, ""))
+
+	err := mountAndRun(t, VCMeetingEvents, []string{
+		"+meeting-events",
+		"--meeting-id", "7628568141510692381",
+		"--format", "pretty",
+		"--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	reg.Verify(t)
+
+	if !strings.Contains(stdout.String(), "No meeting events.") {
+		t.Fatalf("unexpected output: %s", stdout.String())
+	}
+}
+
+func TestParseFlexibleTime(t *testing.T) {
+	t.Run("unix seconds", func(t *testing.T) {
+		got, ok := parseFlexibleTime("1776410100")
+		if !ok {
+			t.Fatal("parseFlexibleTime() ok = false, want true")
+		}
+		if want := time.Unix(1776410100, 0); !got.Equal(want) {
+			t.Fatalf("parseFlexibleTime() = %v, want %v", got, want)
+		}
+	})
+
+	t.Run("unix millis", func(t *testing.T) {
+		got, ok := parseFlexibleTime("1776408061000")
+		if !ok {
+			t.Fatal("parseFlexibleTime() ok = false, want true")
+		}
+		if want := time.UnixMilli(1776408061000); !got.Equal(want) {
+			t.Fatalf("parseFlexibleTime() = %v, want %v", got, want)
+		}
+	})
+
+	t.Run("rfc3339", func(t *testing.T) {
+		got, ok := parseFlexibleTime("2026-04-17T08:00:00Z")
+		if !ok {
+			t.Fatal("parseFlexibleTime() ok = false, want true")
+		}
+		if want, _ := time.Parse(time.RFC3339, "2026-04-17T08:00:00Z"); !got.Equal(want) {
+			t.Fatalf("parseFlexibleTime() = %v, want %v", got, want)
+		}
+	})
+
+	t.Run("invalid", func(t *testing.T) {
+		if _, ok := parseFlexibleTime("not-a-time"); ok {
+			t.Fatal("parseFlexibleTime() ok = true, want false")
+		}
+	})
+}
+
+func TestFormatMeetingWindow(t *testing.T) {
+	start := time.Unix(1776410100, 0)
+	end := time.Unix(1776413700, 0)
+
+	tests := []struct {
+		name     string
+		start    time.Time
+		hasStart bool
+		end      time.Time
+		hasEnd   bool
+		want     string
+	}{
+		{
+			name:     "ongoing",
+			start:    start,
+			hasStart: true,
+			end:      start,
+			hasEnd:   true,
+			want:     "2026-04-17 15:15:00（进行中）",
+		},
+		{
+			name:     "finished range",
+			start:    start,
+			hasStart: true,
+			end:      end,
+			hasEnd:   true,
+			want:     "2026-04-17 15:15:00 - 2026-04-17 16:15:00",
+		},
+		{
+			name:     "only start",
+			start:    start,
+			hasStart: true,
+			want:     "2026-04-17 15:15:00",
+		},
+		{
+			name:   "only end",
+			end:    end,
+			hasEnd: true,
+			want:   "2026-04-17 16:15:00",
+		},
+		{
+			name: "empty",
+			want: "",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := formatMeetingWindow(tt.start, tt.hasStart, tt.end, tt.hasEnd); got != tt.want {
+				t.Fatalf("formatMeetingWindow() = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestFormatTimelineOffset(t *testing.T) {
+	start := time.Unix(1776410100, 0)
+	later := start.Add(90 * time.Second)
+	earlier := start.Add(-5 * time.Minute)
+
+	tests := []struct {
+		name            string
+		when            time.Time
+		hasWhen         bool
+		meetingStart    time.Time
+		hasMeetingStart bool
+		want            string
+	}{
+		{
+			name:            "with meeting start",
+			when:            later,
+			hasWhen:         true,
+			meetingStart:    start,
+			hasMeetingStart: true,
+			want:            "00:01:30",
+		},
+		{
+			name:            "negative diff clamps to zero",
+			when:            earlier,
+			hasWhen:         true,
+			meetingStart:    start,
+			hasMeetingStart: true,
+			want:            "00:00:00",
+		},
+		{
+			name:    "without meeting start uses wall clock",
+			when:    later,
+			hasWhen: true,
+			want:    "15:16:30",
+		},
+		{
+			name: "missing when",
+			want: "??:??:??",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := formatTimelineOffset(tt.when, tt.hasWhen, tt.meetingStart, tt.hasMeetingStart); got != tt.want {
+				t.Fatalf("formatTimelineOffset() = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestFlattenQueryParams(t *testing.T) {
+	params := larkcore.QueryParams{
+		"one":   []string{"1"},
+		"many":  []string{"2", "3"},
+		"empty": []string{},
+	}
+
+	got := flattenQueryParams(params)
+	want := map[string]interface{}{
+		"one":  "1",
+		"many": []string{"2", "3"},
+	}
+	if !reflect.DeepEqual(got, want) {
+		t.Fatalf("flattenQueryParams() = %#v, want %#v", got, want)
+	}
+}
+
+func TestCompactMeetingPayload_DropsOnlyEmptySlices(t *testing.T) {
+	got := compactMeetingPayload(map[string]interface{}{
+		"empty_items": []interface{}{},
+		"items":       []interface{}{"x"},
+		"zero":        0,
+		"text":        "ok",
+	})
+
+	if _, ok := got["empty_items"]; ok {
+		t.Fatalf("compactMeetingPayload() should drop empty_items: %#v", got)
+	}
+	if !reflect.DeepEqual(got["items"], []interface{}{"x"}) {
+		t.Fatalf("compactMeetingPayload() items = %#v, want %#v", got["items"], []interface{}{"x"})
+	}
+	if got["zero"] != 0 || got["text"] != "ok" {
+		t.Fatalf("compactMeetingPayload() preserved fields mismatch: %#v", got)
+	}
+}
+
+func TestCompactMeetingEvents_IgnoresNonMapsAndCompactsPayload(t *testing.T) {
+	got := compactMeetingEvents([]interface{}{
+		"skip-me",
+		map[string]interface{}{
+			"event_type": "chat_received",
+			"payload": map[string]interface{}{
+				"chat_received_items": []interface{}{"x"},
+				"empty_items":         []interface{}{},
+			},
+		},
+	})
+
+	if len(got) != 1 {
+		t.Fatalf("len(compactMeetingEvents()) = %d, want 1", len(got))
+	}
+	event, _ := got[0].(map[string]interface{})
+	payload := common.GetMap(event, "payload")
+	if _, ok := payload["empty_items"]; ok {
+		t.Fatalf("compactMeetingEvents() should prune empty payload slices: %#v", payload)
+	}
+}
+
+func TestVCShortcuts_RegistersMeetingAgentCommands(t *testing.T) {
+	got := Shortcuts()
+	var commands []string
+	for _, shortcut := range got {
+		commands = append(commands, shortcut.Command)
+	}
+	want := []string{"+search", "+notes", "+recording", "+meeting-join", "+meeting-leave", "+meeting-events"}
+	if !reflect.DeepEqual(commands, want) {
+		t.Fatalf("shortcut commands = %#v, want %#v", commands, want)
+	}
+}
+
+func TestLeaveAction(t *testing.T) {
+	tests := []struct {
+		name string
+		item map[string]interface{}
+		want string
+	}{
+		{name: "meeting ended", item: map[string]interface{}{"leave_reason": 2}, want: "因会议结束离开了会议"},
+		{name: "kicked", item: map[string]interface{}{"leave_reason": 3}, want: "被移出了会议"},
+		{name: "default", item: map[string]interface{}{"leave_reason": 1}, want: "离开了会议"},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := leaveAction(tt.item); got != tt.want {
+				t.Fatalf("leaveAction() = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestMeetingEventUserWithID(t *testing.T) {
+	tests := []struct {
+		name string
+		user map[string]interface{}
+		want string
+	}{
+		{name: "nil", want: ""},
+		{name: "name and id", user: map[string]interface{}{"user_name": "Alice", "id": "u1"}, want: "Alice(u1)"},
+		{name: "name only", user: map[string]interface{}{"user_name": "Alice"}, want: "Alice"},
+		{name: "id only", user: map[string]interface{}{"id": "u1"}, want: "u1"},
+		{name: "empty", user: map[string]interface{}{}, want: ""},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := meetingEventUserWithID(tt.user); got != tt.want {
+				t.Fatalf("meetingEventUserWithID() = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestMeetingEventSummary(t *testing.T) {
+	tests := []struct {
+		name  string
+		event map[string]interface{}
+		want  string
+	}{
+		{
+			name: "participant joined count",
+			event: map[string]interface{}{
+				"event_type": "participant_joined",
+				"payload": map[string]interface{}{
+					"participant_joined_items": []interface{}{
+						map[string]interface{}{},
+						map[string]interface{}{},
+					},
+				},
+			},
+			want: "2 participants joined",
+		},
+		{
+			name: "participant left with label",
+			event: map[string]interface{}{
+				"event_type": "participant_left",
+				"payload": map[string]interface{}{
+					"participant_left_items": []interface{}{
+						map[string]interface{}{"participant": map[string]interface{}{"user_name": "Bob", "id": "u2"}},
+					},
+				},
+			},
+			want: "participant u2 (Bob) left",
+		},
+		{
+			name: "fallback unknown event",
+			event: map[string]interface{}{
+				"event_type": "mystery_event",
+				"payload":    map[string]interface{}{},
+			},
+			want: "mystery_event",
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := meetingEventSummary(tt.event); got != tt.want {
+				t.Fatalf("meetingEventSummary() = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestEscapePrettyText(t *testing.T) {
+	got := escapePrettyText("line1\nline2\t\r" + string(rune(0x07)))
+	want := `line1\nline2\t\r\u0007`
+	if got != want {
+		t.Fatalf("escapePrettyText() = %q, want %q", got, want)
+	}
+}
+
+func TestNeedsColon(t *testing.T) {
+	tests := []struct {
+		description string
+		want        bool
+	}{
+		{description: "发送了消息", want: false},
+		{description: "加入了会议", want: false},
+		{description: "离开了会议", want: false},
+		{description: "开始共享「文档」", want: false},
+		{description: "[text] hello", want: true},
+	}
+	for _, tt := range tests {
+		if got := needsColon(tt.description); got != tt.want {
+			t.Fatalf("needsColon(%q) = %v, want %v", tt.description, got, tt.want)
+		}
+	}
+}

shortcuts/vc/vc_meeting_join.go

@@ -0,0 +1,94 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package vc
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"regexp"
+	"strings"
+
+	"github.com/larksuite/cli/shortcuts/common"
+)
+
+var meetingNumberRe = regexp.MustCompile(`^\d{9}$`)
+
+// validMeetingNumber checks whether s is a valid 9-digit meeting number.
+func validMeetingNumber(s string) bool {
+	return meetingNumberRe.MatchString(s)
+}
+
+// VCMeetingJoin joins a meeting by meeting number via /vc/v1/bots/join.
+var VCMeetingJoin = common.Shortcut{
+	Service:     "vc",
+	Command:     "+meeting-join",
+	Description: "Join a meeting by meeting number (bot join)",
+	Risk:        "write",
+	Scopes:      []string{"vc:meeting.bot.join:write"},
+	AuthTypes:   []string{"user"},
+	HasFormat:   true,
+	Flags: []common.Flag{
+		{Name: "meeting-number", Required: true, Desc: "meeting number to join"},
+		{Name: "password", Desc: "meeting password (if required)"},
+	},
+	Validate: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		mn := strings.TrimSpace(runtime.Str("meeting-number"))
+		if !validMeetingNumber(mn) {
+			return common.FlagErrorf("--meeting-number must be exactly 9 digits, got %q", mn)
+		}
+		return nil
+	},
+	DryRun: func(ctx context.Context, runtime *common.RuntimeContext) *common.DryRunAPI {
+		body := buildMeetingJoinBody(runtime)
+		return common.NewDryRunAPI().
+			POST("/open-apis/vc/v1/bots/join").
+			Body(body)
+	},
+	Execute: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		body := buildMeetingJoinBody(runtime)
+		data, err := runtime.DoAPIJSON("POST", "/open-apis/vc/v1/bots/join", nil, body)
+		if err != nil {
+			return err
+		}
+		if data == nil {
+			data = map[string]interface{}{}
+		}
+		runtime.OutFormat(data, nil, func(w io.Writer) {
+			meeting, _ := data["meeting"].(map[string]interface{})
+			if meeting == nil {
+				fmt.Fprintln(w, "Joined meeting (no meeting info returned).")
+				return
+			}
+			fmt.Fprintf(w, "Joined meeting successfully.\n")
+			if id := common.GetString(meeting, "id"); id != "" {
+				fmt.Fprintf(w, "  Meeting ID:  %s\n", id)
+			}
+			if no := common.GetString(meeting, "meeting_no"); no != "" {
+				fmt.Fprintf(w, "  Meeting No:  %s\n", no)
+			}
+			if topic := common.GetString(meeting, "topic"); topic != "" {
+				fmt.Fprintf(w, "  Topic:       %s\n", topic)
+			}
+			if startTime := common.GetString(meeting, "start_time"); startTime != "" {
+				fmt.Fprintf(w, "  Start Time:  %s\n", startTime)
+			}
+		})
+		return nil
+	},
+}
+
+func buildMeetingJoinBody(runtime *common.RuntimeContext) map[string]interface{} {
+	meetingNo := strings.TrimSpace(runtime.Str("meeting-number"))
+	body := map[string]interface{}{
+		"join_type": 1,
+		"join_identify": map[string]interface{}{
+			"meeting_no": meetingNo,
+		},
+	}
+	if pw := strings.TrimSpace(runtime.Str("password")); pw != "" {
+		body["password"] = pw
+	}
+	return body
+}

shortcuts/vc/vc_meeting_leave.go

@@ -0,0 +1,57 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package vc
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"strings"
+
+	"github.com/larksuite/cli/shortcuts/common"
+)
+
+// VCMeetingLeave leaves a meeting via /vc/v1/bots/leave.
+var VCMeetingLeave = common.Shortcut{
+	Service:     "vc",
+	Command:     "+meeting-leave",
+	Description: "Leave a meeting by meeting ID",
+	Risk:        "write",
+	Scopes:      []string{"vc:meeting.bot.join:write"},
+	AuthTypes:   []string{"user"},
+	HasFormat:   true,
+	Flags: []common.Flag{
+		{Name: "meeting-id", Required: true, Desc: "meeting ID to leave"},
+	},
+	Validate: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		if strings.TrimSpace(runtime.Str("meeting-id")) == "" {
+			return common.FlagErrorf("--meeting-id is required")
+		}
+		return nil
+	},
+	DryRun: func(ctx context.Context, runtime *common.RuntimeContext) *common.DryRunAPI {
+		return common.NewDryRunAPI().
+			POST("/open-apis/vc/v1/bots/leave").
+			Body(map[string]interface{}{
+				"meeting_id": strings.TrimSpace(runtime.Str("meeting-id")),
+			})
+	},
+	Execute: func(ctx context.Context, runtime *common.RuntimeContext) error {
+		meetingID := strings.TrimSpace(runtime.Str("meeting-id"))
+		body := map[string]interface{}{
+			"meeting_id": meetingID,
+		}
+		data, err := runtime.DoAPIJSON("POST", "/open-apis/vc/v1/bots/leave", nil, body)
+		if err != nil {
+			return err
+		}
+		if data == nil {
+			data = map[string]interface{}{}
+		}
+		runtime.OutFormat(data, nil, func(w io.Writer) {
+			fmt.Fprintf(w, "Left meeting %s successfully.\n", meetingID)
+		})
+		return nil
+	},
+}

shortcuts/vc/vc_meeting_test.go

@@ -0,0 +1,536 @@
+// Copyright (c) 2026 Lark Technologies Pte. Ltd.
+// SPDX-License-Identifier: MIT
+
+package vc
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"strings"
+	"testing"
+
+	"github.com/spf13/cobra"
+
+	"github.com/larksuite/cli/internal/cmdutil"
+	"github.com/larksuite/cli/internal/httpmock"
+	"github.com/larksuite/cli/shortcuts/common"
+)
+
+// ---------------------------------------------------------------------------
+// Unit tests: pure functions
+// ---------------------------------------------------------------------------
+
+func TestValidMeetingNumber(t *testing.T) {
+	tests := []struct {
+		name string
+		in   string
+		want bool
+	}{
+		{"9 digits", "123456789", true},
+		{"9 digits leading zero", "012345678", true},
+		{"empty", "", false},
+		{"8 digits", "12345678", false},
+		{"10 digits", "1234567890", false},
+		{"with space", "12345 678", false},
+		{"letters mixed", "12345678a", false},
+		{"pure letters", "abcdefghi", false},
+		{"with dash", "123-456-789", false},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := validMeetingNumber(tt.in); got != tt.want {
+				t.Errorf("validMeetingNumber(%q) = %v, want %v", tt.in, got, tt.want)
+			}
+		})
+	}
+}
+
+func TestBuildMeetingJoinBody_WithoutPassword(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-number", "", "")
+	cmd.Flags().String("password", "", "")
+	_ = cmd.Flags().Set("meeting-number", "123456789")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	body := buildMeetingJoinBody(runtime)
+
+	if body["join_type"] != 1 {
+		t.Errorf("join_type = %v, want 1", body["join_type"])
+	}
+	ji, ok := body["join_identify"].(map[string]interface{})
+	if !ok {
+		t.Fatalf("join_identify missing or wrong type: %v", body["join_identify"])
+	}
+	if ji["meeting_no"] != "123456789" {
+		t.Errorf("meeting_no = %v, want 123456789", ji["meeting_no"])
+	}
+	if _, exists := body["password"]; exists {
+		t.Errorf("password should be omitted when empty, got %v", body["password"])
+	}
+}
+
+func TestBuildMeetingJoinBody_WithPassword(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-number", "", "")
+	cmd.Flags().String("password", "", "")
+	_ = cmd.Flags().Set("meeting-number", "123456789")
+	_ = cmd.Flags().Set("password", "secret")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	body := buildMeetingJoinBody(runtime)
+
+	if body["password"] != "secret" {
+		t.Errorf("password = %v, want secret", body["password"])
+	}
+}
+
+func TestBuildMeetingJoinBody_TrimsWhitespace(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-number", "", "")
+	cmd.Flags().String("password", "", "")
+	_ = cmd.Flags().Set("meeting-number", "  123456789  ")
+	_ = cmd.Flags().Set("password", "  pw  ")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	body := buildMeetingJoinBody(runtime)
+
+	ji, _ := body["join_identify"].(map[string]interface{})
+	if ji["meeting_no"] != "123456789" {
+		t.Errorf("meeting_no should be trimmed, got %q", ji["meeting_no"])
+	}
+	if body["password"] != "pw" {
+		t.Errorf("password should be trimmed, got %q", body["password"])
+	}
+}
+
+// ---------------------------------------------------------------------------
+// Validate tests: VCMeetingJoin
+// ---------------------------------------------------------------------------
+
+func TestMeetingJoin_Validate_MissingNumber(t *testing.T) {
+	f, _, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	// cobra MarkFlagRequired should reject missing --meeting-number
+	err := mountAndRun(t, VCMeetingJoin, []string{"+meeting-join", "--as", "user"}, f, nil)
+	if err == nil {
+		t.Fatal("expected error when --meeting-number is missing")
+	}
+	if !strings.Contains(err.Error(), "meeting-number") {
+		t.Errorf("error should mention meeting-number, got: %v", err)
+	}
+}
+
+func TestMeetingJoin_Validate_InvalidFormat(t *testing.T) {
+	tests := []struct {
+		name string
+		num  string
+	}{
+		{"too short", "12345678"},
+		{"too long", "1234567890"},
+		{"with letters", "12345abcd"},
+		{"empty after trim", "   "},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			cmd := &cobra.Command{Use: "test"}
+			cmd.Flags().String("meeting-number", "", "")
+			cmd.Flags().String("password", "", "")
+			_ = cmd.Flags().Set("meeting-number", tt.num)
+
+			runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+			err := VCMeetingJoin.Validate(context.Background(), runtime)
+			if err == nil {
+				t.Fatalf("expected validation error for %q", tt.num)
+			}
+			if !strings.Contains(err.Error(), "9 digits") {
+				t.Errorf("error should mention '9 digits', got: %v", err)
+			}
+		})
+	}
+}
+
+func TestMeetingJoin_Validate_Valid(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-number", "", "")
+	cmd.Flags().String("password", "", "")
+	_ = cmd.Flags().Set("meeting-number", "123456789")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	if err := VCMeetingJoin.Validate(context.Background(), runtime); err != nil {
+		t.Errorf("unexpected validation error: %v", err)
+	}
+}
+
+// ---------------------------------------------------------------------------
+// DryRun tests: VCMeetingJoin
+// ---------------------------------------------------------------------------
+
+func TestMeetingJoin_DryRun(t *testing.T) {
+	f, stdout, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "123456789", "--password", "pw123",
+		"--dry-run", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	out := stdout.String()
+	if !strings.Contains(out, "/open-apis/vc/v1/bots/join") {
+		t.Errorf("dry-run should include API path, got: %s", out)
+	}
+	if !strings.Contains(out, "123456789") {
+		t.Errorf("dry-run should include meeting number, got: %s", out)
+	}
+	if !strings.Contains(out, "pw123") {
+		t.Errorf("dry-run should include password, got: %s", out)
+	}
+}
+
+// ---------------------------------------------------------------------------
+// Execute tests: VCMeetingJoin
+// ---------------------------------------------------------------------------
+
+func TestMeetingJoin_Execute_Success(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	stub := &httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/join",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{
+				"meeting": map[string]interface{}{
+					"id":         "69999999",
+					"meeting_no": "123456789",
+					"topic":      "Weekly Sync",
+					"start_time": "1700000000",
+				},
+			},
+		},
+	}
+	reg.Register(stub)
+
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "123456789",
+		"--format", "json", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	// verify captured request body
+	if len(stub.CapturedBody) == 0 {
+		t.Fatal("expected request body to be captured")
+	}
+	var req map[string]interface{}
+	if err := json.Unmarshal(stub.CapturedBody, &req); err != nil {
+		t.Fatalf("failed to parse request body: %v", err)
+	}
+	if req["join_type"].(float64) != 1 {
+		t.Errorf("join_type = %v, want 1", req["join_type"])
+	}
+	ji, _ := req["join_identify"].(map[string]interface{})
+	if ji["meeting_no"] != "123456789" {
+		t.Errorf("meeting_no = %v, want 123456789", ji["meeting_no"])
+	}
+	if _, exists := ji["password"]; exists {
+		t.Errorf("password should be omitted when not provided, got %v", ji["password"])
+	}
+
+	// verify response envelope carries meeting info under data.meeting
+	var resp map[string]any
+	if err := json.Unmarshal(stdout.Bytes(), &resp); err != nil {
+		t.Fatalf("failed to parse stdout: %v", err)
+	}
+	data, _ := resp["data"].(map[string]any)
+	meeting, _ := data["meeting"].(map[string]any)
+	if meeting["id"] != "69999999" {
+		t.Errorf("meeting.id = %v, want 69999999 (envelope: %s)", meeting["id"], stdout.String())
+	}
+}
+
+func TestMeetingJoin_Execute_WithPassword_CapturesBody(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	stub := &httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/join",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{},
+		},
+	}
+	reg.Register(stub)
+
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "987654321", "--password", "s3cret",
+		"--format", "json", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	var req map[string]interface{}
+	if err := json.Unmarshal(stub.CapturedBody, &req); err != nil {
+		t.Fatalf("failed to parse request body: %v", err)
+	}
+	ji, _ := req["join_identify"].(map[string]interface{})
+	if req["password"] != "s3cret" {
+		t.Errorf("password = %v, want s3cret", req["password"])
+	}
+	if ji["meeting_no"] != "987654321" {
+		t.Errorf("meeting_no = %v, want 987654321", ji["meeting_no"])
+	}
+}
+
+func TestMeetingJoin_Execute_PrettyOutput(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/join",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{
+				"meeting": map[string]interface{}{
+					"id":         "69999999",
+					"meeting_no": "123456789",
+					"topic":      "Weekly Sync",
+					"start_time": "1700000000",
+				},
+			},
+		},
+	})
+
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "123456789",
+		"--format", "pretty", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	out := stdout.String()
+	for _, want := range []string{"Joined meeting successfully", "69999999", "123456789", "Weekly Sync", "1700000000"} {
+		if !strings.Contains(out, want) {
+			t.Errorf("pretty output missing %q, got: %s", want, out)
+		}
+	}
+}
+
+func TestMeetingJoin_Execute_PrettyOutput_NoMeetingInfo(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/join",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{},
+		},
+	})
+
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "123456789",
+		"--format", "pretty", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if !strings.Contains(stdout.String(), "no meeting info returned") {
+		t.Errorf("pretty output should fall back to 'no meeting info' notice, got: %s", stdout.String())
+	}
+}
+
+func TestMeetingLeave_Execute_PrettyOutput(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/leave",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{},
+		},
+	})
+
+	err := mountAndRun(t, VCMeetingLeave, []string{
+		"+meeting-leave", "--meeting-id", "69999999",
+		"--format", "pretty", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	out := stdout.String()
+	if !strings.Contains(out, "Left meeting 69999999 successfully") {
+		t.Errorf("pretty output should confirm leave, got: %s", out)
+	}
+}
+
+func TestMeetingJoin_Execute_APIError(t *testing.T) {
+	f, _, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/join",
+		Body:   map[string]interface{}{"code": 190001, "msg": "invalid meeting number"},
+	})
+
+	err := mountAndRun(t, VCMeetingJoin, []string{
+		"+meeting-join", "--meeting-number", "123456789",
+		"--as", "user",
+	}, f, &bytes.Buffer{})
+	if err == nil {
+		t.Fatal("expected error for API failure")
+	}
+	if !strings.Contains(err.Error(), "invalid meeting number") {
+		t.Errorf("error should surface API message, got: %v", err)
+	}
+}
+
+// ---------------------------------------------------------------------------
+// Validate tests: VCMeetingLeave
+// ---------------------------------------------------------------------------
+
+func TestMeetingLeave_Validate_MissingID(t *testing.T) {
+	f, _, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	err := mountAndRun(t, VCMeetingLeave, []string{"+meeting-leave", "--as", "user"}, f, nil)
+	if err == nil {
+		t.Fatal("expected error when --meeting-id is missing")
+	}
+	if !strings.Contains(err.Error(), "meeting-id") {
+		t.Errorf("error should mention meeting-id, got: %v", err)
+	}
+}
+
+func TestMeetingLeave_Validate_WhitespaceOnly(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-id", "", "")
+	_ = cmd.Flags().Set("meeting-id", "   ")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	err := VCMeetingLeave.Validate(context.Background(), runtime)
+	if err == nil {
+		t.Fatal("expected error for whitespace-only meeting-id")
+	}
+	if !strings.Contains(err.Error(), "meeting-id") {
+		t.Errorf("error should mention meeting-id, got: %v", err)
+	}
+}
+
+func TestMeetingLeave_Validate_Valid(t *testing.T) {
+	cmd := &cobra.Command{Use: "test"}
+	cmd.Flags().String("meeting-id", "", "")
+	_ = cmd.Flags().Set("meeting-id", "69999999")
+
+	runtime := common.TestNewRuntimeContext(cmd, defaultConfig())
+	if err := VCMeetingLeave.Validate(context.Background(), runtime); err != nil {
+		t.Errorf("unexpected validation error: %v", err)
+	}
+}
+
+// ---------------------------------------------------------------------------
+// DryRun tests: VCMeetingLeave
+// ---------------------------------------------------------------------------
+
+func TestMeetingLeave_DryRun(t *testing.T) {
+	f, stdout, _, _ := cmdutil.TestFactory(t, defaultConfig())
+	err := mountAndRun(t, VCMeetingLeave, []string{
+		"+meeting-leave", "--meeting-id", "69999999",
+		"--dry-run", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	out := stdout.String()
+	if !strings.Contains(out, "/open-apis/vc/v1/bots/leave") {
+		t.Errorf("dry-run should include API path, got: %s", out)
+	}
+	if !strings.Contains(out, "69999999") {
+		t.Errorf("dry-run should include meeting-id, got: %s", out)
+	}
+}
+
+// ---------------------------------------------------------------------------
+// Execute tests: VCMeetingLeave
+// ---------------------------------------------------------------------------
+
+func TestMeetingLeave_Execute_Success(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	stub := &httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/leave",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{},
+		},
+	}
+	reg.Register(stub)
+
+	err := mountAndRun(t, VCMeetingLeave, []string{
+		"+meeting-leave", "--meeting-id", "69999999",
+		"--format", "json", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	// verify captured request body
+	var req map[string]interface{}
+	if err := json.Unmarshal(stub.CapturedBody, &req); err != nil {
+		t.Fatalf("failed to parse request body: %v", err)
+	}
+	if req["meeting_id"] != "69999999" {
+		t.Errorf("meeting_id = %v, want 69999999", req["meeting_id"])
+	}
+}
+
+func TestMeetingLeave_Execute_TrimsMeetingID(t *testing.T) {
+	f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	stub := &httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/leave",
+		Body: map[string]interface{}{
+			"code": 0, "msg": "ok",
+			"data": map[string]interface{}{},
+		},
+	}
+	reg.Register(stub)
+
+	err := mountAndRun(t, VCMeetingLeave, []string{
+		"+meeting-leave", "--meeting-id", "  69999999  ",
+		"--format", "json", "--as", "user",
+	}, f, stdout)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	var req map[string]interface{}
+	if err := json.Unmarshal(stub.CapturedBody, &req); err != nil {
+		t.Fatalf("failed to parse request body: %v", err)
+	}
+	if req["meeting_id"] != "69999999" {
+		t.Errorf("meeting_id should be trimmed, got %q", req["meeting_id"])
+	}
+}
+
+func TestMeetingLeave_Execute_APIError(t *testing.T) {
+	f, _, _, reg := cmdutil.TestFactory(t, defaultConfig())
+
+	reg.Register(&httpmock.Stub{
+		Method: "POST",
+		URL:    "/open-apis/vc/v1/bots/leave",
+		Body:   map[string]interface{}{"code": 121005, "msg": "no permission"},
+	})
+
+	err := mountAndRun(t, VCMeetingLeave, []string{
+		"+meeting-leave", "--meeting-id", "69999999", "--as", "user",
+	}, f, &bytes.Buffer{})
+	if err == nil {
+		t.Fatal("expected error for API failure")
+	}
+	if !strings.Contains(err.Error(), "no permission") {
+		t.Errorf("error should surface API message, got: %v", err)
+	}
+}

skills/lark-base/SKILL.md

@@ -12,6 +12,7 @@ metadata:
 
 > **前置条件：** 先阅读 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)。
 > **执行前必做：** 执行任何 `base` 命令前，必须先阅读对应命令的 reference 文档，再调用命令。
+> **查询类任务必做：** 涉及筛选、排序、Top/Bottom N、聚合、多表关联、查询后写入或判断全局结论时，必须先阅读 [`references/lark-base-data-analysis-sop.md`](references/lark-base-data-analysis-sop.md)，再选择 `record / view / data-query` 路径。
 > **命名约定：** Base 业务命令仅使用 `lark-cli base +...` 形式；如需先解析 Wiki 链接，可先调用 `lark-cli wiki ...`。
 > **分流规则：** 如果用户要“把本地文件导入成 Base / 多维表格 / bitable”，第一步不是 `base`，而是 `lark-cli drive +import --type bitable`；导入完成后再回到 `lark-cli base +...` 做表内操作。
 
@@ -104,7 +105,7 @@ metadata:
 
 | 命令 | 用途 / 何时使用 | 必读 reference | 路由提醒 |
 |------|------------------|----------------|----------|
-| `+record-search / +record-list / +record-get` | 按关键词检索记录、读取记录明细 / 分页导出，或按 ID 获取一条或多条记录 | [`lark-base-record-read-sop.md`](references/lark-base-record-read-sop.md) | 记录读取统一先读 read SOP guide：已知 `record_id` 用 `+record-get`；明确关键词用 `+record-search`；普通明细用 `+record-list`；明确筛选 / 排序 / Top N 用临时视图投影后 `+record-list --view-id`；统计聚合才分流到 `+data-query`；`+record-get` 支持重复 `--record-id` 或 `--json` 读取多条记录 |
+| `+record-search / +record-list / +record-get` | 按关键词检索记录、读取记录明细 / 分页导出，或按 ID 获取一条或多条记录 | [`lark-base-data-analysis-sop.md`](references/lark-base-data-analysis-sop.md) | 记录读取统一先读 data analysis SOP：已知 `record_id` 用 `+record-get`；明确关键词用 `+record-search`；普通明细用 `+record-list`；明确筛选 / 排序 / Top N 用临时视图投影后 `+record-list --view-id`；统计聚合才分流到 `+data-query`；`+record-get` 支持重复 `--record-id` 或 `--json` 读取多条记录 |
 | `+record-upsert / +record-batch-create / +record-batch-update` | 创建、更新或批量写入记录 | [`lark-base-record-upsert.md`](references/lark-base-record-upsert.md)、[`lark-base-record-batch-create.md`](references/lark-base-record-batch-create.md)、[`lark-base-record-batch-update.md`](references/lark-base-record-batch-update.md)、[`lark-base-cell-value.md`](references/lark-base-cell-value.md) | 写前先 `+field-list`；只写存储字段；`+record-batch-update` 为同值更新（同一 patch 应用到多条记录）；批量单次不超过 `200` 条；附件不要走这里 |
 | `+record-upload-attachment` | 给已有记录上传附件 | [`lark-base-record-upload-attachment.md`](references/lark-base-record-upload-attachment.md) | 附件上传专用链路，不要用 `+record-upsert` / `+record-batch-*` 伪造附件值 |
 | `lark-cli docs +media-download` | 下载 Base 附件文件到本地 | [`../lark-doc/references/lark-doc-media-download.md`](../lark-doc/references/lark-doc-media-download.md) | Base 附件的 `file_token` 从 `+record-get` 返回的附件字段数组里取；**不要用 `lark-cli drive +download`**（对 Base 附件返回 403） |
@@ -120,7 +121,7 @@ metadata:
 |------|------------------|----------------|----------|
 | `+view-list / +view-get` | 列出视图，或获取视图详情 | [`lark-base-view-list.md`](references/lark-base-view-list.md)、[`lark-base-view-get.md`](references/lark-base-view-get.md) | `+view-list` 只能串行执行；`+view-get` 适合查看已有视图配置 |
 | `+view-create / +view-delete / +view-rename` | 创建、删除或重命名视图 | [`lark-base-view-create.md`](references/lark-base-view-create.md)、[`lark-base-view-delete.md`](references/lark-base-view-delete.md)、[`lark-base-view-rename.md`](references/lark-base-view-rename.md) | 创建前先确认表和视图类型；删除前先确认目标；用户已明确新名字时可直接重命名 |
-| `+view-get-filter / +view-set-filter` | 读取或配置筛选条件 | [`lark-base-view-get-filter.md`](references/lark-base-view-get-filter.md)、[`lark-base-view-set-filter.md`](references/lark-base-view-set-filter.md)、[`lark-base-record-read-sop.md`](references/lark-base-record-read-sop.md) | 常与 `+record-list` 组合，用于按视图筛选读取 |
+| `+view-get-filter / +view-set-filter` | 读取或配置筛选条件 | [`lark-base-view-get-filter.md`](references/lark-base-view-get-filter.md)、[`lark-base-view-set-filter.md`](references/lark-base-view-set-filter.md)、[`lark-base-data-analysis-sop.md`](references/lark-base-data-analysis-sop.md) | 常与 `+record-list` 组合，用于按视图筛选读取 |
 | `+view-get-sort / +view-set-sort` | 读取或配置排序 | [`lark-base-view-get-sort.md`](references/lark-base-view-get-sort.md)、[`lark-base-view-set-sort.md`](references/lark-base-view-set-sort.md) | 字段名必须来自真实结构 |
 | `+view-get-group / +view-set-group` | 读取或配置分组 | [`lark-base-view-get-group.md`](references/lark-base-view-get-group.md)、[`lark-base-view-set-group.md`](references/lark-base-view-set-group.md) | 字段名必须来自真实结构 |
 | `+view-get-visible-fields / +view-set-visible-fields` | 读取或配置视图可见字段 | [`lark-base-view-get-visible-fields.md`](references/lark-base-view-get-visible-fields.md)、[`lark-base-view-set-visible-fields.md`](references/lark-base-view-set-visible-fields.md) | 用于控制视图中的字段顺序与可见性；字段名必须来自真实结构 |
@@ -228,14 +229,24 @@ metadata:
 | 基于视图做筛选读取 | `+view-set-filter` + `+record-list` | 不要跳过视图筛选直接猜条件 |
 | 本地 Excel / CSV / `.base` 导入为 Base | `lark-cli drive +import --type bitable` | 不要误走 `+base-create`、`+table-create` 或 `+record-upsert` |
 
-### 3.3 表名、字段名与表达式引用
+### 3.3 查询执行契约
+
+涉及查询、统计或判断结论时，先阅读 [`references/lark-base-data-analysis-sop.md`](references/lark-base-data-analysis-sop.md)，并遵守以下高优先级规则：
+
+1. `+record-list` 默认页、固定 `--limit` 和本地 `jq` 只能证明已读取范围内的事实，不能直接支撑全局最值、全量计数、Top/Bottom N、异常识别或分组结论。
+2. 能由 Base 表达的筛选、排序、投影、聚合、分组和限制，应在 Base 云端查询服务中执行；不要先拉明细到本地上下文再手工筛选排序。
+3. `has_more=true` 或等价分页信号表示当前结果不是全量；除非用户只要样例/前 N 条，不能基于该页回答全局问题。
+4. 多表查询必须先确认关系字段和连接键；link 单元格里的 `record_id` 是关系键，不是用户可读答案。
+5. 最终答案必须能追溯到真实表、真实字段、查询范围、筛选/排序/聚合条件和必要的连接键。
+
+### 3.4 表名、字段名与表达式引用
 
 1. 表名、字段名必须精确匹配真实返回，来源应是 `+table-list / +table-get / +field-list`。
 2. 不要凭自然语言猜名称，不要自行改写用户口述中的表名、字段名。
 3. `formula / lookup / data-query / workflow` 中出现的名称同样必须精确匹配；表达式引用、where 条件、DSL 字段名、workflow 配置都遵守同一规则。
 4. 跨表场景必须额外读取目标表结构，不能只看当前表。
 
-### 3.4 Token 与链接
+### 3.5 Token 与链接
 
 这是高优先级章节。只要用户输入里出现链接、token，或报错涉及 `baseToken` / `wiki_token` / `obj_token`，都应优先回到这里检查。
 
@@ -254,7 +265,7 @@ metadata:
 | `slides` | 转到 Drive 相关 skill | 不继续使用本 skill 的 Base 命令 |
 | `mindnote` | 转到 Drive 相关 skill | 不继续使用本 skill 的 Base 命令 |
 
-### 3.5 身份选择与权限降级策略
+### 3.6 身份选择与权限降级策略
 
 多维表格通常属于用户的个人或团队资源。**默认应优先使用 `--as user`（用户身份）执行所有 Base 操作**，始终显式指定身份。
 
@@ -282,10 +293,11 @@ lark-cli auth login --domain base
 
 1. 先判断任务属于哪个模块，选对命令族。
 2. 如果用户给了链接，先解析 token，不要把 wiki token、完整 URL 或其他对象 ID 误当成 `base_token`。
-3. 先拿结构，再写命令，避免猜表名、字段名、表达式引用。
-4. 定位到命令后，先读对应 reference，再执行命令。
-5. 执行命令，并按返回结果判断下一步。
-6. 回复时返回关键结果和后续可继续操作的信息，方便 agent 链式执行下一步。
+3. 如果是查询类任务，先判断问题范围，阅读 data analysis SOP，再决定使用 `record / view / data-query`。
+4. 先拿结构，再写命令，避免猜表名、字段名、表达式引用。
+5. 定位到命令后，先读对应 reference，再执行命令。
+6. 执行命令，并按返回结果判断下一步。
+7. 回复时返回关键结果和后续可继续操作的信息，方便 agent 链式执行下一步。
 
 ### 4.2 不可违反规则
 
@@ -297,11 +309,12 @@ lark-cli auth login --domain base
 6. 只写可写字段；系统字段、附件字段、`formula`、`lookup` 默认不作为普通记录写入目标。
 7. 聚合分析与取数分流；统计走 `+data-query`，关键词检索走 `+record-search`，明细走 `+record-list / +record-get`。
 8. 筛选查询按视图能力执行；先用 `+view-set-filter` 配置筛选，再结合 `+record-list` 读取。
-9. Base 场景不要改走裸 API，不要切去 `lark-cli api /open-apis/bitable/v1/...`。
-10. 统一使用 `--base-token`。
-11. workflow 场景先读 schema，不要凭自然语言猜 `type`。
-12. dashboard 场景先读 guide；提到图表、看板、block 就先进入 dashboard 模块。
-13. formula / lookup 场景先读 guide；没读 guide 前不要直接创建或更新。
+9. 全局查询不得基于默认分页、小 `--limit` 或未证明全量的本地 `jq` 结果下结论。
+10. Base 场景不要改走裸 API，不要切去 `lark-cli api /open-apis/bitable/v1/...`。
+11. 统一使用 `--base-token`。
+12. workflow 场景先读 schema，不要凭自然语言猜 `type`。
+13. dashboard 场景先读 guide；提到图表、看板、block 就先进入 dashboard 模块。
+14. formula / lookup 场景先读 guide；没读 guide 前不要直接创建或更新。
 
 ### 4.3 并发、分页与批量限制
 

skills/lark-base/references/lark-base-data-analysis-sop.md

@@ -0,0 +1,88 @@
+# Base data analysis SOP
+
+Base 数据查询与分析任务的执行契约。覆盖记录读取、筛选、排序、Top/Bottom N、聚合统计、分组聚合、多表关联、临时分析和查询后写入前的目标定位。
+
+具体命令参数不要在本文猜；需要时跳到对应 reference：
+
+- `+data-query`: [lark-base-data-query.md](lark-base-data-query.md)
+- 视图筛选/排序/投影: [lark-base-view-set-filter.md](lark-base-view-set-filter.md), [lark-base-view-set-sort.md](lark-base-view-set-sort.md), [lark-base-view-set-visible-fields.md](lark-base-view-set-visible-fields.md)
+- 记录读取: [lark-base-record.md](lark-base-record.md)
+
+## 0. Hard Rules
+
+- 全局问题不能用默认 `+record-list --limit N` 片面地回答。
+- `jq` / shell / 本地代码是在个人电脑或当前运行环境中处理已返回数据，只适合小范围结果；超过 200 行默认不推荐本地统计、排序或求极值，应改用 Base 云端查询服务的 filter/sort/aggregate。
+- “最高、最低、最新、最早、Top、Bottom、总数、全部、异常、最大、最小、最多、最少、优先级最高”等全局语义，必须在 Base 云端查询服务中完成筛选、排序或聚合。
+- `+record-search` 用于关键词检索字段的展示文本；可搜多类字段，但匹配的是文本表示（如人员命中 name），不要用它替代金额、状态、日期、空值等结构化条件。
+- 不要依赖已有视图，除非用户明确指定该视图，或你已读取并验证其 filter/sort/projection 符合当前问题。
+- 交付输出必须使用用户可读的真实字段值；内部 ID、`record_id`、关联记录 ID、open_id、编码字段只可作为连接键或定位键，不能替代最终输出，除非用户明确要求输出这些键值。
+- 每次读取必须做最小投影，并包含后续解释、回查或写入需要的业务 key。
+
+## 1. Intent -> Tool Path
+
+| 用户意图 | 首选路径 | 关键规则 |
+| --- | --- | --- |
+| 看几条、预览、示例 | `+record-list --limit N` | 保持局部语义；不要推广为全局结论 |
+| 已知 `record_id` | `+record-get` | 直接读取；不要 search/list 反查 |
+| 明确关键词 | `+record-search` | 按字段展示文本命中；使用 `search_fields` 限定匹配范围、`select_fields` 投影降低返回内容 token 量；不要把文本检索当作结构化关联解析 |
+| 按条件找明细记录 | 先创建临时视图设置筛选和可见字段，再用 `+record-list --view-id` 读取 | 条件字段来自 `+field-list`；不要先读全表再本地过滤 |
+| 排序 / TopN 原始记录 | 临时视图 filter/sort/projection -> `+record-list --view-id --limit N` | 最高/最新降序，最低/最早升序 |
+| 聚合 / 分组 / 分组排序 | `+data-query` | 使用 filters/dimensions/measures/sort/limit |
+| 聚合后输出实体字段 | `+data-query` 得到业务 key -> record 路径回查明细 | `+data-query` 不返回原始记录或 link 明细；聚合结果中的 key 需要再解析成用户要求字段 |
+| 多表 / 多跳关联 | 以候选数最小的事实表为驱动表，沿业务 key 或 link `record_id` 逐跳回查 | 读出 link 单元格里的关联 `record_id` 后，到被关联表批量 `+record-get` 展示字段，并在回答用结果中合并展示 |
+| 查询后写入 / 视图化 | 先用本 SOP 得到可复核的目标记录 id 集合 | 再进入记录写入或视图配置；高价值可复用查询优先沉淀为持久视图 |
+
+## 2. Execution Patterns
+
+### 2.1 结构化明细与 TopN
+
+使用视图路径：
+
+1. `+field-list` 确认筛选字段、排序字段、展示字段、业务 key。
+2. `+view-create` 创建 grid 视图。
+3. 设置 filter/sort/visible fields。
+4. `+record-list --view-id <view_id> --limit <N>` 读取结果。
+
+不要从未筛选、未排序的全表输出中手动挑选。一次性查询可用临时视图；如果这个筛选/排序结果对用户后续查看有价值，应保留为持久视图，不要删除，并告知用户视图名称和用途。视图参数细节见 view-set references。
+
+### 2.2 聚合分析与 TopN
+
+使用 `+data-query`：
+
+- 让 Base 云端查询服务完成 filters、dimensions、measures、sort、pagination.limit。
+- `pagination.limit` 是 Base 云端查询服务中的聚合结果限制，不是本地分页扫描。
+- 需要输出明细或用户可读字段时，先拿业务 key，再用 record 路径精确回查。
+- 字段类型、日期 value、DSL shape 以 [lark-base-data-query.md](lark-base-data-query.md) 为准。
+
+### 2.3 关系查询与回查
+
+- link 单元格通常是关联表 `record_id` 数组，不是用户可读内容，只是连接键。
+- 先用 `+field-list` 确认 link 字段的 `link_table`、业务唯一键和展示字段。
+- 从驱动表拿到候选记录后，用关联 `record_id` 到关联表 `+record-get` 批量读取记录内容。
+- 多跳关系逐跳建立 `record_id/key -> 用户可读字段` 映射；最终用户可读的信息。
+
+禁止：
+
+- 把 link `record_id` 当最终输出。
+- 用 `+record-search` 搜 link `record_id`。
+- 基于 ID、自增编号、link 值做语义猜测；禁止依赖字段先验、样本记忆补全交付输出。
+
+## 3. Range & Pagination Contract
+
+- `+record-list` 默认页、固定 `--limit`、本地 `jq`、shell 管道、手工浏览输出，都只覆盖已读取范围；超过 200 行不要把本地处理当作推荐路径。
+- `has_more=true`、存在下一页 offset/page token、或返回行数等于 page size，都表示可能还有未读取数据。
+- 对全局问题，只有 Base 云端查询服务已经通过 filter/sort/aggregate 收敛目标范围，或 `data-query` 已在云端完成聚合、排序和限制时，才可以用有限返回形成结论。
+- 必须全量导出时，按 CLI 分页语义串行翻页；不要并发调用 `+record-list`。
+
+## 4. Final Answer Check
+
+形成交付输出前必须能确认：
+
+- 问题范围是局部样例、单点定位、全局明细、聚合分析、多表关联，还是查询后写入。
+- 筛选、排序、聚合是否发生在 Base 云端查询服务中，而不是本地 `jq` / shell 中。
+- 如果使用 `jq` / shell，本地输入是否是 200 行以内的小范围结果；超过 200 行是否已改用 Base 云端查询服务查询。
+- 如果使用 `+record-list`，是否处理了 `has_more`，且投影包含业务 key 和解释字段。
+- 如果涉及关系查询，是否按 `record_id` 或业务 key 精确回查，交付输出是否来自关联表真实字段。
+- 交付输出能追溯到表、字段、筛选条件、排序/聚合条件和连接键。
+
+任一项无法确认时，继续查询或明确说明只能得到局部结论。

skills/lark-base/references/lark-base-data-query.md

@@ -5,6 +5,8 @@
 
 对多维表格数据进行聚合查询（分组、过滤、排序、聚合计算），基于以下语法的 JSON DSL：
 
+查询类任务还必须先遵守 [`lark-base-data-analysis-sop.md`](lark-base-data-analysis-sop.md)。`+data-query` 适合让筛选、分组、聚合、排序和 TopN 在 Base 云端查询服务中执行；不要用默认分页的 `+record-list` 或本地 `jq` 替代聚合查询。
+
 ## 限制
 
 - **权限要求**（按文档类型分流）：
@@ -51,6 +53,23 @@ lark-cli base +data-query \
     "measures": [{"field_name": "金额", "aggregation": "sum", "alias": "total"}],
     "shaper": {"format": "flat"}
   }'
+
+# 聚合后如需读取明细，先让 data-query 返回可回查的业务 key
+lark-cli base +data-query \
+  --base-token MAGObxxxxx \
+  --dsl '{
+    "datasource": {"type": "table", "table": {"tableId": "tblxxxxxxxx"}},
+    "dimensions": [{"field_name": "业务编号", "alias": "biz_key"}],
+    "measures": [{"field_name": "指标值", "aggregation": "max", "alias": "max_value"}],
+    "filters": {
+      "type": 1,
+      "conjunction": "and",
+      "conditions": [{"field_name": "状态", "operator": "is", "value": ["有效"]}]
+    },
+    "sort": [{"field_name": "max_value", "order": "desc"}],
+    "pagination": {"limit": 10},
+    "shaper": {"format": "flat"}
+  }'
 ```
 
 ## 参数
@@ -397,6 +416,19 @@ value 使用预定义关键字机制，第一个元素为字符串常量名称
    - 每个 value 是 CellValue 对象，实际值在 `value` 字段中，如 `{"value": "北京"}` 或 `{"value": 12345.00}`
    - 失败时结果在 `data.error` 中，包含具体错误码和信息
 
+## 与记录读取组合
+
+`+data-query` 不返回原始记录或 link 字段明细。需要输出聚合结果对应的原始记录字段、展示值或关联表字段时，按以下方式组合：
+
+1. 用 `+data-query` 在 Base 云端查询服务中完成全局筛选、分组、聚合、排序和 TopN，得到业务 key、分组值或候选范围。
+2. 如果已经拿到候选记录的 `record_id`，用 `+record-get` 读取明细字段。
+3. 如果拿到的是结构化业务 key（例如编号、状态、日期、金额等），优先创建临时视图做精确过滤后再 `+record-list --view-id` 读取；不要用 `+record-search` 代替结构化条件。
+4. 只有候选条件本身是文本展示值关键词时，才使用 `+record-search`，并用 `search_fields` 限定范围、`select_fields` 做投影。
+5. 若候选记录包含 link 字段，提取关联 `record_id` 后到关联表用 `+record-get` 批量读取展示字段。
+6. 最终回答业务字段，不要把内部 `record_id` 当作用户可读答案。
+
+不要把 `data-query pagination.limit` 理解为分页扫描；它只限制 Base 云端查询服务返回的聚合结果行数，不支持 offset。需要全量明细导出时回到 data analysis SOP 的 record 分页规则。
+
 ## 坑点
 
 - ⚠️ **必须先查表结构**：DSL 的 `field_name` 必须与表中字段名称精确匹配（区分大小写），不能凭猜测构造。先用 `lark-cli base +field-list --base-token <base_token> --table-id <table_id>` 获取真实字段名
@@ -408,10 +440,12 @@ value 使用预定义关键字机制，第一个元素为字符串常量名称
 - ⚠️ **数据表标识 `tableId` vs `tableName`**：datasource 中可以用 `tableId`（如 `tblXXX`）或 `tableName`（数据表的用户自定义显示名称），二选一，不要混用
 - ⚠️ **`pagination.limit` 最大 5000**：超过会报错，且不支持 offset，只支持 limit
 - ⚠️ **所有 alias 必须全局唯一**：dimensions 和 measures 之间的 alias 也不能重名
+- ⚠️ **不要用本地分页结果替代 data-query**：凡是全局计数、分组、聚合、排序 TopN，优先让 `+data-query` 在 Base 云端查询服务中执行；默认页 `+record-list` 后本地统计只能得到已读取范围内的结果
 
 ## 参考
 
 - [lark-base](../SKILL.md) — 多维表格全部命令
 - [lark-shared](../../lark-shared/SKILL.md) — 认证和全局参数
+- [lark-base-data-analysis-sop.md](lark-base-data-analysis-sop.md) — 查询范围、选路、下推、分页、record 明细回查和关系查询 SOP
 - [lark-base-cell-value.md](lark-base-cell-value.md) — CellValue 格式规范
 - [lark-base-shortcut-field-properties.md](lark-base-shortcut-field-properties.md) — shortcut 字段类型与 JSON 结构

skills/lark-base/references/lark-base-record-read-sop.md

@@ -1,86 +0,0 @@
-# base record read SOP
-
-> **前置条件：** 先阅读 [`../lark-shared/SKILL.md`](../../lark-shared/SKILL.md) 了解认证、权限处理和全局参数。
-
-记录读取由 6 个功能组合完成：选路、字段投影、视图预处理、分页与范围、返回结构解释、link 关联读取。
-
-## 1. 读取选路
-
-| 场景 | 使用方式 | 规则 |
-|------|------|------|
-| 已知 `record_id` | `+record-get` | 只读单条记录，不要用 search/list 反查。 |
-| 明确关键词检索 | `+record-search` | 只用于文本关键词检索；金额、状态、日期等结构化条件不要用 search。 |
-| 普通明细读取 / 导出 / 查看前 N 条 | `+record-list` | 优先加 `--view-id` 时只读该视图可见记录与可见字段；或者加 `--field-id` 手动裁剪字段；不传 `--view-id` 时会读取全表。 |
-| 明确筛选 / 排序 / Top N / Bottom N 且需要原始记录或 `record_id` | 创建带 filter + sort 的临时视图 + `+record-list --view-id` | 让视图完成 filter/sort projection，LLM 不擅长手工筛选排序，建议用视图完成。 |
-| 统计 / 聚合结果且不需要 `record_id` | 转到 [`lark-base-data-query.md`](lark-base-data-query.md) | `data-query` 是特殊分析 DSL，不是记录读取工具。 |
-
-## 2. 字段投影
-
-- `FieldListFirst`: 不清楚字段结构时先 `+field-list`，确认筛选字段、排序字段、展示字段、关联字段、业务唯一键字段。
-- `UseRealField`: 字段名和字段 ID 必须来自 `+field-list` 返回，不要凭自然语言猜字段名。
-- `MinimalProjection`: 每次读取只返回本次任务需要的字段；`+record-list` 用重复 `--field-id`，视图读取用 `+view-set-visible-fields`。
-- `FieldScopePriority`: 返回字段优先级为显式投影字段（`+record-list --field-id` / `record-search select_fields`） > 视图可见字段 > 全表字段；需要稳定列范围时必须显式投影。
-- `LongFieldAvoidance`: 默认不要读取 `trace`、`raw`、长文本、附件等高噪声字段，除非任务明确需要。
-- `BusinessKey`: 后续要定位、更新或解释记录时，投影中必须包含可识别业务字段，例如订单号、日报ID、姓名、编号。
-
-## 3. 视图预处理
-
-适用于结构化筛选、排序、最高/最低、倒数、Top/Bottom N、按条件找记录等场景。
-
-1. `+field-list` 获取字段 ID、字段名和字段类型。
-2. `+view-create` 创建临时 `grid` 视图，名称带任务语义，例如 `tmp_query_销售额升序`。
-3. `+view-set-filter` 设置筛选条件；空值是否参与必须按用户语义判断。
-4. `+view-set-sort` 设置排序条件；最高/最新用降序，最低/最早/倒数用升序。
-5. `+view-set-visible-fields` 设置投影字段，只保留业务键、排序字段、筛选解释字段、需要展示或二跳的字段。
-6. `+record-list --view-id <view_id> --limit <N>` 读取结果；不要再从未排序全表输出中手动挑选。
-
-## 4. 分页与范围
-
-- `ViewScope`: URL 带 `view_id` 时先判断用户是否要求“该视图下”；全表问题不要误用 URL 视图范围，应该根据需求创建合适的临时视图完成查询任务。
-- `ViewIdScope`: `+record-list --view-id` 是作用域参数；仅用于用户指定的视图，或本次任务主动创建的临时筛选 / 排序 / 投影视图。
-- `NeedAllPages`: 用户要求全部、导出、统计、最高/最低且未用视图/limit 限定时，必须检查 `has_more` 并串行翻页。
-- `LimitWhenScoped`: 用户只要示例、前 N 条、Top/Bottom N，使用 `--limit` 控制结果规模。
-- `NoConcurrentList`: `+record-list` 禁止并发调用；分页和多表读取必须串行。
-- `DataQueryScope`: `data-query` 的筛选 DSL 与视图筛选不是同一套语法；不要混用。
-
-## 5. 返回结构解释
-
-- `ColumnMapping`: `fields` / `field_id_list` 定义 `data` 每列含义；解释记录前先建立列到字段名的映射。
-- `RowMapping`: `record_id_list[i]` 与 `data[i]` 是同一行；需要后续定位、更新或关联时，按下标整理成 `record_id + 字段名:值` 的小表。
-- `BusinessMatch`: 后续引用目标记录时按业务字段匹配，不靠肉眼数行号。
-- `FieldType`: 按字段类型解释值；数字、货币、日期、人员、formula、lookup、attachment、link 不要当普通文本处理。
-- `EmptyValue`: 空值参与筛选或排序前必须明确语义；不要默认把空值当 `0`、空字符串或有效状态。
-- `AnswerCheck`: 最终回答前复核答案记录来自读取结果、筛选排序已应用、字段含义和 record_id 映射无误。
-
-## 6. link 关联字段读取
-
-link 字段是关联单元格；读取结果通常是关联表的 `record_id` 数组，不是用户可读名称。
-
-| 步骤 | 做法 |
-|------|------|
-| 识别 link 字段 | 用 `+field-list` 查看字段类型为 `link`，并读取 `link_table` 确认关联目标表。 |
-| 读取当前表 | 在当前表 `+record-list` / `+record-get` 中保留 link 字段和业务键字段。 |
-| 解析单元格值 | link 单元格通常形如 `[{"id":"rec..."}]`；提取其中每个 `id` 作为关联表 `record_id`。 |
-| 读取关联表 | 到 `link_table` 使用 `+record-get --record-id <rec...>` 或裁剪后的 `+record-list` 读取显示字段。 |
-| 建立映射 | 形成 `关联record_id -> 显示字段值` 映射，再回填当前表结果。 |
-| 多值处理 | 多个关联值保持原顺序；可去重批量读取，但回答时按原单元格顺序输出。 |
-
-禁止事项：
-
-- 不要把 link 单元格里的 `record_id` 当作最终答案。
-- 不要用 `+record-search` 搜索 link `record_id` 来查关联记录。
-- 不要凭关联 `record_id` 猜名称、负责人、门店等显示值。
-- 不要只看当前表字段名推断关联表结构；跨表读取前必须拿关联表字段结构。
-
-## 7. 命令 help
-
-- `HelpFirst`: 参数、示例、JSON shape 和取值约束以 `lark-cli base +record-get --help`、`+record-search --help`、`+record-list --help` 为准。
-- `RecordSearchJson`: 构造 `+record-search --json` 前先看 `+record-search --help`，确认 `keyword/search_fields/select_fields/view_id/offset/limit` 的结构和约束。
-- `RecordListProjection`: 构造 `+record-list` 前先看 `+record-list --help`，确认 `--field-id`、`--view-id`、`--offset`、`--limit` 的语义。
-
-## 参考
-
-- [lark-base-view-set-filter.md](lark-base-view-set-filter.md)
-- [lark-base-view-set-sort.md](lark-base-view-set-sort.md)
-- [lark-base-view-set-visible-fields.md](lark-base-view-set-visible-fields.md)
-- [lark-base-data-query.md](lark-base-data-query.md)

skills/lark-base/references/lark-base-record.md

@@ -8,7 +8,7 @@ record 相关命令索引。
 
 | 文档 | 命令 | 说明 |
 |------|------|------|
-| [lark-base-record-read-sop.md](lark-base-record-read-sop.md) | `+record-get` / `+record-search` / `+record-list` | 记录读取统一选路、筛选排序投影 SOP |
+| [lark-base-data-analysis-sop.md](lark-base-data-analysis-sop.md) | `+record-get` / `+record-search` / `+record-list` / `+data-query` / 视图筛选排序 | 数据查询与分析统一选路、筛选排序投影、聚合后回查明细 SOP |
 | [lark-base-record-upsert.md](lark-base-record-upsert.md) | `+record-upsert` | 创建或更新记录 |
 | [lark-base-record-batch-create.md](lark-base-record-batch-create.md) | `+record-batch-create` | 按 `fields/rows` 批量创建记录 |
 | [lark-base-record-batch-update.md](lark-base-record-batch-update.md) | `+record-batch-update` | 批量更新记录 |
@@ -19,7 +19,7 @@ record 相关命令索引。
 
 ## 说明
 
-- 读取记录前优先阅读 [lark-base-record-read-sop.md](lark-base-record-read-sop.md)，它合并了 `+record-get` / `+record-search` / `+record-list` 的选路和 SOP。
+- 读取记录前优先阅读 [lark-base-data-analysis-sop.md](lark-base-data-analysis-sop.md)，它合并了 `record / view / data-query` 的选路、分页、投影、聚合后回查明细和 link 关联读取。
 - 聚合页只保留目录职责；写入、删除、历史等命令的详细说明请进入对应单命令文档。
 - 所有 `+xxx-list` 调用都必须串行执行；若要批量跑多个 list 请求，只能串行执行。
 - `+record-list` 支持重复传参 `--field-id` 做字段筛选。

skills/lark-shared/SKILL.md

@@ -12,7 +12,9 @@ description: "飞书/Lark CLI 共享基础：应用配置初始化、认证登
 
 首次使用需运行 `lark-cli config init` 完成应用配置。
 
-当你帮用户初始化配置时，使用background方式使用下面的命令发起配置应用流程，启动后读取输出，从中提取授权链接并发给用户：
+当你帮用户初始化配置时，使用background方式使用下面的命令发起配置应用流程，启动后读取输出，从中提取授权链接并发给用户。
+
+**URL 转发规则**：当命令输出 `verification_url`、`verification_uri_complete`、`console_url` 等 URL 字段时，必须将 URL exactly as returned by the CLI 转发给用户，并把它视为不可修改的 opaque string；不要做 URL encode/decode，不要补 `%20`、空格或标点，不要重新拼接 query，不要改写成 Markdown link text，建议用只包含原始 URL 的代码块单独输出。
 
 ```bash
 # 发起配置（该命令会阻塞直到用户打开链接并完成操作或过期）
@@ -51,7 +53,7 @@ lark-cli config init --new
 
 #### Bot 身份（`--as bot`）
 
-将错误中的 `console_url` 提供给用户，引导去后台开通 scope。**禁止**对 bot 执行 `auth login`。
+将错误中的 `console_url` 原样提供给用户，引导去后台开通 scope。**禁止**对 bot 执行 `auth login`。
 
 #### User 身份（`--as user`）
 
@@ -64,7 +66,7 @@ lark-cli auth login --scope "<missing_scope>"   # 按具体 scope 授权（推
 
 #### Agent 代理发起认证（推荐）
 
-当你作为 AI agent 需要帮用户完成认证时，使用background方式 执行以下命令发起授权流程, 并将授权链接发给用户：
+当你作为 AI agent 需要帮用户完成认证时，使用background方式 执行以下命令发起授权流程, 并将授权链接原样发给用户：
 
 ```bash
 # 发起授权（阻塞直到用户授权完成或过期）

skills/lark-slides-creator/SKILL.md

@@ -0,0 +1,107 @@
+---
+name: lark-slides-creator
+version: 1.0.0
+description: "飞书幻灯片创作工作流：从自然语言需求创建、重构、美化完整 PPT，覆盖规划、模板选择、视觉风格、素材规划和创建后验证。"
+metadata:
+  requires:
+    bins: ["lark-cli"]
+  cliHelp: "lark-cli slides --help"
+---
+
+# slides creator workflow
+
+> 执行 XML/API 前必须读取 ../lark-slides/SKILL.md 和对应 reference。
+
+This skill is the natural-language entry point for creating polished presentations. It owns planning, design, template, asset, and quality-validation workflows. It delegates all XML/API execution to [`../lark-slides/SKILL.md`](../lark-slides/SKILL.md).
+
+## When To Use
+
+Use this skill when the user asks for:
+
+- A new complete presentation from a topic, notes, outline, document, meeting, or rough prompt.
+- Beautification, restructuring, major rewrite, or formal-report polishing.
+- Template selection or a deck based on a theme, scene, industry, or visual style.
+- Visual direction, palette, typography, layout system, or executive-ready presentation quality.
+- Asset planning, image search/download/upload planning, or deciding where visuals belong.
+- Creation-time and post-creation validation for content completeness and visual quality.
+
+For a narrow raw XML/API operation, use `lark-slides` directly.
+
+## Required Execution Dependency
+
+Before running any `lark-cli slides` command or writing final XML:
+
+1. Read [`../lark-slides/SKILL.md`](../lark-slides/SKILL.md).
+2. Read [`../lark-slides/references/xml-schema-quick-ref.md`](../lark-slides/references/xml-schema-quick-ref.md).
+3. Read the relevant execution reference, such as `lark-slides-create.md`, `lark-slides-media-upload.md`, `lark-slides-replace-slide.md`, or an `xml_presentation.*` API reference.
+
+Use the execution skill's lint tool from here when XML is available:
+
+```bash
+python3 skills/lark-slides-creator/scripts/layout_lint.py --input /tmp/presentation.xml
+```
+
+## Workflow
+
+1. Understand the deck goal.
+   Capture topic, audience, page count, source material, language, formality, delivery setting, and any brand/style constraints. If the user gives enough information, proceed with explicit assumptions instead of blocking on questions.
+
+2. Choose template or custom direction.
+   If the request mentions templates, style, theme, or a common deck scenario, search templates first:
+
+   ```bash
+   python3 skills/lark-slides-creator/scripts/template_tool.py search --query "<用户需求原文>" --limit 3
+   ```
+
+   Offer 2-3 concise candidates when user choice matters. If one template is clearly best for a lightweight request, state the default and continue unless the user asked to choose.
+
+3. Plan the deck.
+   Build a page-by-page outline with title, role, key message, and intended layout for each slide. For formal reports, make the argument flow explicit: context, evidence, analysis, recommendation, next steps.
+
+4. Design the visual system.
+   Define palette, typography hierarchy, spacing, page rhythm, chart/table treatment, and recurring elements. Keep slides visual and low-density; do not produce document-like pages.
+
+5. Plan assets.
+   Decide which pages need screenshots, photos, diagrams, icons, or charts. External images must become local files first, then execution uses `+media-upload` or `@./path` placeholders as described in `lark-slides`.
+
+6. Generate XML and execute through `lark-slides`.
+   Use template summaries or extracted page slices when helpful, but rewrite all placeholder copy into the user's real content. For complex decks, prefer the two-step create flow from `lark-slides`.
+
+7. Validate after creation.
+   Read the created presentation XML with `xml_presentations get`, confirm page count and expected content, run lint when possible, then fix issues with `+replace-slide` or raw slide APIs.
+
+## Template Workflow
+
+Template assets live in this skill:
+
+- [`references/template-catalog.md`](references/template-catalog.md)
+- [`references/template-index.json`](references/template-index.json)
+- [`assets/templates/`](assets/templates/)
+- [`scripts/template_tool.py`](scripts/template_tool.py)
+
+Machine-first commands:
+
+```bash
+python3 skills/lark-slides-creator/scripts/template_tool.py search --query "工作汇报" --tone light --limit 3
+python3 skills/lark-slides-creator/scripts/template_tool.py summarize --template office--work_report --label 内容
+python3 skills/lark-slides-creator/scripts/template_tool.py extract --template office--work_report --label 封面 --out /tmp/work-report-cover.xml
+```
+
+Rules:
+
+- Search using the user's original wording.
+- Show only 2-3 candidate templates unless the user asks for the full catalog.
+- Summarize a target page type before extracting XML.
+- Do not read entire template XML files by default.
+- Reuse theme, spacing, and structure; do not copy placeholder text.
+
+## References
+
+| Reference | Purpose |
+| --- | --- |
+| [planning-layer.md](references/planning-layer.md) | Deck planning and outline workflow. |
+| [visual-planning.md](references/visual-planning.md) | Visual style and layout design guidance. |
+| [asset-planning.md](references/asset-planning.md) | Asset selection, local-file, and upload planning. |
+| [template-catalog.md](references/template-catalog.md) | Template matching catalog. |
+| [slide-templates.md](references/slide-templates.md) | Copyable slide XML patterns for creation. |
+| [validation-checklist.md](references/validation-checklist.md) | Creation quality and post-create validation checklist. |

skills/lark-slides-creator/references/asset-planning.md

@@ -0,0 +1,21 @@
+# Asset Planning
+
+Use this when a deck needs screenshots, photos, diagrams, logos, icons, or chart data.
+
+## Asset Plan
+
+For each asset, record:
+
+- Slide number and purpose.
+- Asset type: screenshot, product image, chart, diagram, logo, icon, photo.
+- Source: provided file, generated file, downloaded file, or chart from data.
+- Local path under the current working directory.
+- Intended placement and dimensions.
+
+## Rules
+
+- Slides XML cannot use HTTP(S) image URLs directly.
+- For a new deck using `+create --slides`, local image placeholders can use `src="@./path.png"`.
+- For existing decks or raw slide APIs, upload first with `slides +media-upload`, then use the returned `file_token`.
+- Keep source files inside the current working directory or a safe project subdirectory.
+- Check image dimensions and file size before upload; slides media upload limit is 20 MB.

skills/lark-slides-creator/references/planning-layer.md

@@ -0,0 +1,32 @@
+# Slides Planning Layer
+
+Use this before writing XML for a full presentation or a major rewrite.
+
+## Inputs
+
+- Goal: what decision, update, teaching outcome, or story the deck must support.
+- Audience: executives, customers, internal team, interview panel, students, or general readers.
+- Constraints: page count, language, source material, deadline, brand rules, required sections.
+- Success criteria: what the user should be able to inspect after creation.
+
+## Output Outline
+
+Use this compact structure:
+
+```text
+Title: <deck title>
+Audience: <audience>
+Style: <visual direction or selected template>
+Slides:
+1. <slide title> - <message> - <layout intent>
+2. ...
+```
+
+For formal reports, prefer this flow: cover, context, key findings, supporting evidence, implications, recommendations, next steps, closing.
+
+## Rules
+
+- Each slide gets one primary message.
+- Avoid document-like density; split overloaded pages.
+- Make charts or tables serve a stated point.
+- Confirm template choice when multiple good candidates would lead to materially different decks.

skills/lark-slides-creator/references/template-catalog.md

@@ -10,26 +10,26 @@
 
 ## 使用方法
 
-1. 先运行 `python3 skills/lark-slides/scripts/template_tool.py search --query "<主题>" --limit 3`，根据用户描述的 **场景、风格、色调** 做初筛
+1. 先运行 `python3 skills/lark-slides-creator/scripts/template_tool.py search --query "<主题>" --limit 3`，根据用户描述的 **场景、风格、色调** 做初筛
 2. 整理出 **2-3 个**最匹配的用户可选模板候选；优先选场景强相关模板，没有明显场景模板时再用标 ⭐ 的通用模板兜底
 3. 用户选定后，再锁定 **1-2 个**最匹配的模板作为实际参考
 4. 先看模板下方的 **页型索引**，锁定你真正需要的页型：封面 / 目录 / 分节 / 内容 / 结尾
 5. 优先运行 `template_tool.py summarize` 查看 `<theme>` / 页型摘要；只有需要具体布局骨架时，再运行 `template_tool.py extract`
 6. 从模板中提取并复用：`<theme>` 配色、页面流、shape 排列布局、装饰元素风格
 7. 将用户的实际内容填充到模板的结构框架中，**不要照搬模板的占位文字**
-8. 创建前运行 `layout_lint.py --input <file>`；它检查 XML well-formed 和布局风险，不等价于完整 XSD schema 校验
+8. 创建前运行 `python3 skills/lark-slides-creator/scripts/layout_lint.py --input <file>`；它检查 XML well-formed 和布局风险，不等价于完整 XSD schema 校验
 
 ### 脚本快捷命令
 
 ```bash
 # 先找候选模板
-python3 skills/lark-slides/scripts/template_tool.py search --query "工作汇报" --tone light --limit 3
+python3 skills/lark-slides-creator/scripts/template_tool.py search --query "工作汇报" --tone light --limit 3
 
 # 看指定页型的紧凑摘要
-python3 skills/lark-slides/scripts/template_tool.py summarize --template office--work_report --label 内容
+python3 skills/lark-slides-creator/scripts/template_tool.py summarize --template office--work_report --label 内容
 
 # 只裁切目标页型，避免把整份 XML 拉进上下文
-python3 skills/lark-slides/scripts/template_tool.py extract --template office--work_report --label 封面 --out /tmp/work-report-cover.xml
+python3 skills/lark-slides-creator/scripts/template_tool.py extract --template office--work_report --label 封面 --out /tmp/work-report-cover.xml
 ```
 
 如果脚本路径不可用，按这个顺序手动降级：

skills/lark-slides-creator/references/validation-checklist.md

@@ -0,0 +1,25 @@
+# Slides Creation Validation Checklist
+
+Use this after generating XML and again after creating or editing the deck.
+
+## Before API Execution
+
+- XML is well-formed.
+- User text is escaped: `&`, `<`, and `>` are safe.
+- Each slide has one clear message.
+- Text boxes are sized for expected content.
+- Images use `@./local-path` only where `+create --slides` supports it; otherwise they use `file_token`.
+- Run execution-layer lint when XML is in a file:
+
+```bash
+python3 skills/lark-slides-creator/scripts/layout_lint.py --input /tmp/presentation.xml
+```
+
+## After Creation
+
+- Record `xml_presentation_id`.
+- Read the full deck with `xml_presentations get`.
+- Confirm expected page count and page order.
+- Confirm key titles, body text, metrics, and image elements exist.
+- Check for blank pages, missing text, truncated shell arguments, unresolved `@` paths, and wrong image `src`.
+- Fix localized issues with `+replace-slide`; only delete/recreate a page when the whole structure is wrong.

skills/lark-slides-creator/references/visual-planning.md

@@ -0,0 +1,22 @@
+# Visual Planning
+
+Use this to define the deck's visual system before generating slide XML.
+
+## Decisions
+
+- Palette: background, primary accent, secondary accent, text, muted text, border.
+- Typography: title, section heading, body, caption, metric number.
+- Layout rhythm: margins, grid, recurring title position, footer treatment.
+- Components: cards, callouts, timelines, charts, tables, quote blocks, section dividers.
+
+## Guidance
+
+- Business reports should be quiet, readable, and scannable.
+- Product or technology decks can use stronger contrast, but keep hierarchy clear.
+- Use repeated structure across related slides.
+- Keep text inside predictable bounds; leave enough whitespace for rendering variance.
+- Do not rely on external image URLs in XML. Images must become `file_token` values through the execution workflow.
+
+## XML Note
+
+Before writing XML, read `../lark-slides/references/xml-schema-quick-ref.md`. Gradient fills must use `rgba()` stops with percentages.

skills/lark-slides/SKILL.md

@@ -1,525 +1,163 @@
 ---
 name: lark-slides
 version: 1.0.0
-description: "飞书幻灯片：创建和编辑幻灯片，接口通过 XML 协议通信。创建演示文稿、读取幻灯片内容、管理幻灯片页面（创建、删除、读取、局部替换）。当用户需要创建或编辑幻灯片、读取或修改单个页面时使用。"
+description: "飞书幻灯片执行层：通过 Slides XML/API 读取、创建、删除、替换幻灯片页面，处理 URL/wiki token、媒体上传、XML schema、格式校验与排障。"
 metadata:
   requires:
     bins: ["lark-cli"]
   cliHelp: "lark-cli slides --help"
 ---
 
-# slides (v1)
+# slides execution layer
 
-**CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理**
+> 创建完整 PPT、设计、美化、模板、素材、正式汇报场景请使用 lark-slides-creator。本 skill 只负责 XML/API 执行层。
 
-**CRITICAL — 生成任何 XML 之前，MUST 先用 Read 工具读取 [xml-schema-quick-ref.md](references/xml-schema-quick-ref.md)，禁止凭记忆猜测 XML 结构。**
+**CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理。**
 
-**CRITICAL — 如果用户提到“模板”“套用模板”“参考某种主题/风格/版式”，或用户需求明显落在已有场景模板内（如工作汇报、产品介绍、商业计划书、培训、晋升汇报等），MUST 先用 [`scripts/template_tool.py`](scripts/template_tool.py) 的 `search` 做模板检索；默认给出 2-3 个最匹配模板候选供用户选择。锁定模板后用 `summarize` 获取主题和布局摘要；只有需要布局骨架时才用 `extract` 裁切目标页型 XML。不要直接读取完整模板 XML。**
+**CRITICAL — 生成或修改任何 XML 之前，MUST 先读取 [xml-schema-quick-ref.md](references/xml-schema-quick-ref.md)。不要凭记忆猜测 XML 结构。**
 
-> [!NOTE]
-> `scripts/template_tool.py` 需要 Python 3。`references/template-index.json` 是脚本缓存/轻量路由索引，不是默认给 agent 阅读的文档；`assets/templates/*.xml` 是机器资源，只应通过脚本摘要或裁切，不要全文读取。
+**CRITICAL — `references/slides_xml_schema_definition.xml` 是 Slides XML 协议的唯一权威来源；Markdown reference 只是摘要。若两者或 `lark-cli schema` 输出不一致，以 schema 和 CLI 为准。**
 
-**CRITICAL — 使用模板生成或改写页面时，MUST 先 `summarize` 目标页型；只有需要具体布局骨架时才 `extract`。生成本地 XML 后，如可运行 Python，MUST 先用 [`scripts/layout_lint.py`](scripts/layout_lint.py) 检查 XML well-formed、重叠/越界/文本高度风险，再创建或追加页面。它不是完整 XSD schema 校验。**
+## Scope
 
-**编辑已有幻灯片页面**：优先用 [`+replace-slide`](references/lark-slides-replace-slide.md)（块级替换/插入，不动页序）；选择 action 和完整读-改-写流程见 [`lark-slides-edit-workflows.md`](references/lark-slides-edit-workflows.md)。
+Use this skill for low-level execution tasks:
 
-## 身份选择
+- Create an empty presentation or add raw slide XML.
+- Read presentation or slide XML.
+- Delete slides.
+- Replace or insert existing slide blocks.
+- Upload local media and use returned `file_token` in XML.
+- Resolve `/slides/` URL tokens and `/wiki/` tokens.
+- Check XML format, schema rules, and common API errors.
 
-飞书幻灯片通常是用户自己的内容资源。**默认应优先显式使用 `--as user`（用户身份）执行 slides 相关操作**，始终显式指定身份。
+Do not use this skill as the primary entry for planning, visual design, template selection, asset planning, or full-deck creation. Route those requests to `lark-slides-creator`, then return here only for XML/API execution.
 
-- **`--as user`（推荐）**：以当前登录用户身份创建、读取、管理演示文稿。执行前先完成用户授权：
+## Identity
+
+Slides are usually user-owned content. Default to explicit `--as user` for slides commands.
 
 ```bash
 lark-cli auth login --domain slides
 ```
 
-- **`--as bot`**：仅在用户明确要求以应用身份操作，或需要让 bot 持有/创建资源时使用。使用 bot 身份时，要额外确认 bot 是否真的有目标演示文稿的访问权限。
+Use `--as bot` only when the user explicitly asks for app/bot identity or the workflow intentionally creates bot-owned resources. If access fails, first check that the command did not accidentally use the wrong identity.
 
-**执行规则**：
+## URL And Wiki Tokens
 
-1. 创建、读取、增删 slide、按用户给出的链接继续编辑已有 PPT，默认都先用 `--as user`。
-2. 如果出现权限不足，先检查当前是否误用了 bot 身份；不要默认回退到 bot。
-3. 只有在用户明确要求"用应用身份 / bot 身份操作"，或当前工作流就是 bot 创建资源后再做协作授权时，才切换到 `--as bot`。
-
-## 快速开始
-
-一条命令创建包含页面内容的 PPT（推荐）：
+| URL | Token | Handling |
+| --- | --- | --- |
+| `/slides/<token>` | `xml_presentation_id` | Use the path token directly. |
+| `/wiki/<token>` | `wiki_token` | Resolve first with `wiki.spaces.get_node`; use `node.obj_token` only when `node.obj_type` is `slides`. |
 
 ```bash
-lark-cli slides +create --title "演示文稿标题" --slides '[
-  "<slide xmlns=\"http://www.larkoffice.com/sml/2.0\"><style><fill><fillColor color=\"rgb(245,245,245)\"/></fill></style><data><shape type=\"text\" topLeftX=\"80\" topLeftY=\"80\" width=\"800\" height=\"100\"><content textType=\"title\"><p>页面标题</p></content></shape><shape type=\"text\" topLeftX=\"80\" topLeftY=\"200\" width=\"800\" height=\"200\"><content textType=\"body\"><p>正文内容</p><ul><li><p>要点一</p></li><li><p>要点二</p></li></ul></content></shape></data></slide>"
+lark-cli wiki spaces get_node --as user --params '{"token":"wiki_token"}'
+lark-cli slides xml_presentations get --as user --params '{"xml_presentation_id":"obj_token"}'
+```
+
+`+replace-slide` and `+media-upload` can parse slides/wiki URLs. Raw API commands still require the real `xml_presentation_id`.
+
+## Shortcuts
+
+| Shortcut | Reference | Purpose |
+| --- | --- | --- |
+| `slides +create` | [lark-slides-create.md](references/lark-slides-create.md) | Create a presentation; optionally add pages with `--slides`; supports local image placeholders in `+create --slides`. |
+| `slides +media-upload` | [lark-slides-media-upload.md](references/lark-slides-media-upload.md) | Upload a local image to a presentation and return a `file_token`. |
+| `slides +replace-slide` | [lark-slides-replace-slide.md](references/lark-slides-replace-slide.md) | Replace or insert blocks on an existing slide without changing page order. |
+
+Prefer shortcuts when they cover the operation, especially `+replace-slide` for existing-slide edits.
+
+## API Commands
+
+Always inspect schema before raw API calls:
+
+```bash
+lark-cli schema slides.<resource>.<method>
+lark-cli slides <resource> <method> --as user --params '{}' --data '{}'
+```
+
+Core resources:
+
+| Resource | Method | Purpose |
+| --- | --- | --- |
+| `xml_presentations` | `get` | Read full presentation XML and metadata. |
+| `xml_presentation.slide` | `create` | Add one slide XML page. |
+| `xml_presentation.slide` | `delete` | Delete a slide; a presentation must keep at least one page. |
+| `xml_presentation.slide` | `get` | Read one slide XML. |
+| `xml_presentation.slide` | `replace` | Low-level block replace/insert API; prefer `+replace-slide` unless you need raw control. |
+
+## Creation Paths
+
+For simple XML, `+create --slides` is concise:
+
+```bash
+lark-cli slides +create --as user --title "Demo" --slides '[
+  "<slide xmlns=\"http://www.larkoffice.com/sml/2.0\"><style><fill><fillColor color=\"rgb(248,250,252)\"/></fill></style><data><shape type=\"text\" topLeftX=\"80\" topLeftY=\"80\" width=\"800\" height=\"100\"><content textType=\"title\"><p>Title</p></content></shape></data></slide>"
 ]'
 ```
 
-也可以分两步（先创建空白 PPT，再逐页添加），详见 [+create 参考文档](references/lark-slides-create.md)。
-
-> [!WARNING]
-> `--slides '[...]'` 适合简单页面批量创建，但并不等同于“10 页以内都安全”。如果 slide XML 含中文、大段文本、复杂布局、嵌套引号或较多特殊字符，shell 传参时可能出现转义或截断问题，导致内容丢失、页面空白或布局异常。遇到复杂页面时，优先改用“两步创建法”。
-
-> [!IMPORTANT]
-> `slides +create --slides` 底层是“先创建空白 PPT，再逐页调用 `xml_presentation.slide.create`”。这不是原子操作；中途某一页失败时，前面已创建成功的页面会保留。skill 必须把这种“部分成功”风险提前告诉用户，并在失败后先记录 `xml_presentation_id`，回读确认当前状态，再决定是否在现有 PPT 上继续修复或追加。
-
-> 以上是最小可用示例。更丰富的页面效果（渐变背景、卡片、图表、表格等），参考下方 Workflow 和 XML 模板。
-
-## 执行前必做
-
-> **重要**：`references/slides_xml_schema_definition.xml` 是此 skill 唯一正确的 XML 协议来源；其他 md 仅是对它和 CLI schema 的摘要。
-
-### 必读（每次创建前）
-
-| 文档 | 说明 |
-|------|------|
-| [xml-schema-quick-ref.md](references/xml-schema-quick-ref.md) | **XML 元素和属性速查，必读** |
-
-### 选读（需要时查阅）
-
-| 场景 | 文档 |
-|------|------|
-| 需要了解详细 XML 结构 | [xml-format-guide.md](references/xml-format-guide.md) |
-| 需要快速筛模板、做低成本路由 | [`scripts/template_tool.py search`](scripts/template_tool.py) |
-| 需要匹配 PPT 模板/主题风格 | [template-catalog.md](references/template-catalog.md) |
-| 需要按页型抽摘要或裁切 XML 片段 | [`scripts/template_tool.py`](scripts/template_tool.py) |
-| 需要做本地布局风险检查 | [`scripts/layout_lint.py`](scripts/layout_lint.py) |
-| 需要 CLI 调用示例 | [examples.md](references/examples.md) |
-| 需要参考真实 PPT 的 XML | [slides_demo.xml](references/slides_demo.xml) |
-| 需要用 table/chart 等复杂元素 | [slides_xml_schema_definition.xml](references/slides_xml_schema_definition.xml)（完整 Schema） |
-| 需要编辑已有 PPT 的单个页面 | [lark-slides-edit-workflows.md](references/lark-slides-edit-workflows.md) |
-| 需要了解某个命令的详细参数 | 对应命令的 reference 文档（见下方参考文档章节） |
-
-## Workflow
-
-> **这是演示文稿，不是文档。** 每页 slide 是独立的视觉画面，信息密度要低，排版要留白。
-
-### 创建方式选择
-
-| 场景 | 推荐方式 |
-|------|----------|
-| 简单 XML（1-3 页、结构简单、几乎无复杂中文和特殊字符） | `slides +create --slides '[...]'` 一步创建 |
-| 复杂 XML（多页、含中文、大段文本、复杂布局、嵌套引号、特殊字符较多） | **两步创建**：先 `slides +create` 创建空白 PPT，再用 `xml_presentation.slide create` 逐页添加 |
-| 已有 PPT 继续追加或插入页面 | 使用 `xml_presentation.slide create`，必要时配合 `before_slide_id` |
-
-> [!WARNING]
-> `--slides '[...]'` 的风险点主要在 shell 参数传递，而不是单纯页数。即使只有 1 页，只要 XML 足够复杂，也建议使用两步创建法。
-
-### 模板与脚本优先流程
+For complex XML, long text, many special characters, Chinese paragraphs, images, or many pages, create an empty presentation first and add slides one by one. `+create --slides` is not atomic; if a later slide fails, earlier slides may already exist. Record `xml_presentation_id` and read the deck before continuing.
 
 ```bash
-# 1. 搜索候选：把用户原始需求整句放进 --query，不要只放手动提炼的短词
-python3 skills/lark-slides/scripts/template_tool.py search --query "<用户需求原文>" --limit 3
+lark-cli slides +create --as user --title "Demo"
 
-# 2. 锁定模板后先看页型摘要
-python3 skills/lark-slides/scripts/template_tool.py summarize --template <template-id> --label <封面|目录|分节|内容|结尾>
-
-# 3. 只有需要复用布局骨架时才裁切 XML
-python3 skills/lark-slides/scripts/template_tool.py extract --template <template-id> --label <页型> --out /tmp/template-slice.xml
-
-# 4. 生成待创建 XML 后先做布局风险检查
-python3 skills/lark-slides/scripts/layout_lint.py --input /tmp/presentation.xml
-```
-
-执行规则：
-
-1. `search --query` 使用用户原始描述；如用户明确风格，再额外加 `--tone light|dark|colorful` 或 `--formality formal|casual|creative`。
-2. 候选展示只给 2-3 个，包含模板名、适用场景、风格/色调、推荐理由；不要把完整目录贴给用户。
-3. 锁定模板后，复用 `<theme>`、配色、页面流、布局骨架；所有占位文案都必须改写为用户真实内容。
-4. `layout_lint.py` 有 error 时先修 XML，不要提交创建；只有 warning 时，检查是否是可接受的装饰/背景误报。
-
-```text
-Step 1: 需求澄清 & 读取知识
-  - 澄清用户需求：主题、受众、页数、风格偏好
-  - 如果需求明显落在已有模板场景内，主动提示用户“可以直接基于现成模板生成”，并给出 2-3 个最匹配模板候选（模板名 + 适用场景 + 风格/色调 + 简短推荐理由）
-  - 默认不要把完整模板目录直接贴给用户；除非用户明确要求看更多，否则只展示 2-3 个候选
-  - 候选优先选场景强相关模板；只有没有明显场景模板时，才用 `light_general.xml` / `dark_general.xml` 这类通用模板兜底
-  - 如果用户没有明确风格，根据主题推荐（见下方风格判断表）
-  - 如果用户要求“模板/主题/风格参考”，或主题属于常见模板场景：
-    · 优先运行 `python3 skills/lark-slides/scripts/template_tool.py search --query "<用户需求原文>" --limit 3` 做低成本模板匹配
-    · 需要人类可读说明时，再读 template-catalog.md 组织候选文案
-    · 锁定模板后，优先运行 `template_tool.py summarize` 看 `<theme>` / 页型摘要；需要具体布局时，再用 `template_tool.py extract`
-    · 复用模板的 theme、配色、页面流、布局骨架，不要照搬占位文案
-    · `references/template-index.json` 只是脚本缓存/轻量路由索引，`assets/templates/*.xml` 是机器资源；除非用户明确要求审计原始模板，否则不要直接读取
-  - 读取 XML Schema 参考：
-    · xml-schema-quick-ref.md — 元素和属性速查
-    · xml-format-guide.md — 详细结构与示例
-    · slides_demo.xml — 真实 XML 示例
-
-Step 2: 生成大纲 → 用户确认 → 创建
-  - 生成大纲前，先确认用户是否采用推荐模板；轻量任务且候选中有明显最佳匹配时，可在大纲里声明“默认基于 <template-id> 改写”并继续，但正式创建前必须给用户改选机会
-  - 生成结构化大纲（每页标题 + 要点 + 布局描述），交给用户确认
-  - 如果已选模板，大纲和页面布局要明确标注“基于哪个模板/哪些模板改写”
-  - 如果用户明确不要模板，直接按自定义风格继续，不要重复推动模板选择
-  - 先判断创建方式：
-    · 简单 XML：可用 `slides +create --slides '[...]'` 一步创建
-    · 复杂 XML：优先先 `slides +create` 创建空白 PPT，再用 `xml_presentation.slide.create` 逐页添加
-    · 超过 10 页：默认使用两步创建，避免单次输入过长
-  - 含本地图片：
-    · 新建带图 PPT —— 在 slide XML 里写 <img src="@./pic.png" .../>，
-      +create 会自动上传并替换为 file_token（详见 lark-slides-create.md）
-    · 给已有 PPT 加带图新页 —— 先 `slides +media-upload --file ./pic.png --presentation $PID`
-      拿到 file_token，再用它写进 slide XML 调 xml_presentation.slide.create
-    · 给已有页加图 —— 两步：① `slides +media-upload` 拿 file_token
-      ② `slides +replace-slide --parts '[{"action":"block_insert","insertion":"<img src=\"<file_token>\" .../>"}]'`
-      不动其他元素，不要再整页重建（完整示例见 lark-slides-edit-workflows.md 的 block_insert 章节）
-    · 路径必须是 CWD 内的相对路径（如 ./pic.png 或 ./assets/x.png）；
-      绝对路径会被 CLI 拒绝，先 cd 到素材所在目录再执行
-  - 每页 slide 需要完整的 XML：背景、文本、图形、配色
-  - 复杂元素（table、chart）需参考 XSD 原文
-  - 创建前必须做 XML 自检：
-    · 检查特殊字符是否按 XML 规则转义：文本节点和属性值里的裸 `& -> &amp;`；文本里的 `< -> &lt;`、`> -> &gt;`。例如 `Q&A -> Q&amp;A`，URL 属性 `a=1&b=2 -> a=1&amp;b=2`
-    · 属性值里的双引号必须转义或改为外层安全包装，避免 shell 和 JSON 双重截断
-    · 确认所有标签闭合，且 `<slide>` 直接子元素只包含 `<style>`、`<data>`、`<note>`
-    · 如果内容里同时出现中文、大段文本、复杂布局、较多特殊字符，默认不要走 `--slides '[...]'`，直接改用两步创建法
-    · 如果 XML 已落到本地文件且可运行 Python，先执行 `layout_lint.py --input <file>`；它会先检查 XML well-formed 再检查布局风险，但不等价于完整 XSD schema 校验；有 error 先修复再创建
-  - 如果使用模板生成页面，先复用模板骨架再填内容，不要直接复制模板中的长段占位文本
-
-Step 3: 审查 & 交付
-  - 创建完成后，必须用 xml_presentations.get 读取全文 XML 做创建后验证，确认：
-    · 页数是否正确？
-    · 每页 `<data>` 是否包含预期的 `<shape>` / `<img>` / 其他元素？
-    · 文本内容是否完整，是否有被截断、丢失、空白区域？
-    · 关键布局坐标和尺寸是否合理，是否出现明显重叠？
-    · 配色是否统一？字号层级是否合理？
-  - 如果本地有 Python 3，运行
-    `python3 skills/lark-slides/scripts/layout_lint.py --input presentation.xml`
-    做重叠、越界、页脚碰撞、文本高度风险检查；有 error 先修复再交付
-  - 如果创建过程中失败：
-    · 先保留并记录 `xml_presentation_id`，不要假设失败代表什么都没创建
-    · 先判断是否已有部分页面写入，再决定是否在现有 PPT 上修复后继续追加
-    · 优先排查当前失败页：先看该页 XML，再检查是否存在未转义 `&`、错误引号、标签未闭合、shell 传参截断
-  - 局部问题 → 用 `+replace-slide` 块级修正；整页结构要改 → `slide.delete` 旧页 + `slide.create` 新页
-  - 没问题 → 交付：告知用户演示文稿 ID 和访问方式
-```
-
-### 创建后验证
-
-创建成功不等于内容正确。创建完 PPT 后，**必须**读取全文 XML 校验结果：
-
-```bash
-lark-cli slides xml_presentations get --as user \
-  --params '{"xml_presentation_id":"YOUR_ID"}'
-```
-
-重点检查：
-
-- [ ] 页数是否与预期一致
-- [ ] 每页 `<data>` 中是否包含所有预期元素
-- [ ] 文本内容是否完整，没有被 shell 截断或转义损坏
-- [ ] 白底内容区、卡片区、图文区等关键布局是否实际生成
-- [ ] 坐标、宽高是否合理，是否出现堆叠或越界
-
-发现问题时：
-
-1. 不要假设“创建成功就代表渲染正确”
-2. 先读取问题页的 XML，确认是生成问题还是传参损坏
-3. 删除问题页后重新添加；复杂页面优先改用两步创建法
-
-### 最小验收清单
-
-创建完成后，默认按下面顺序验收，不要省略：
-
-1. 记录 `xml_presentation_id`
-2. 确认返回的 `slides_added` 或实际页数是否符合预期
-3. 立即执行 `xml_presentations get`
-4. 检查标题、关键页面、关键文本是否存在
-5. 检查是否有明显空白页、内容缺失、页序错误
-6. 再决定是否向用户交付 URL 和后续编辑建议
-
-推荐最小闭环：
-
-```bash
-# 创建
-lark-cli slides +create --as user --title "Demo" --slides '[...]'
-
-# 立即回读
-lark-cli slides xml_presentations get --as user \
-  --params '{"xml_presentation_id":"YOUR_ID"}'
-```
-
-## XML 自检与排障
-
-在真正创建前，至少做下面 4 项检查：
-
-- [ ] 特殊字符已转义：正文和标题里的 `&`、`<`、`>` 不能裸写；属性值里的裸 `&` 也必须写成 `&amp;`
-- [ ] 属性引号安全：XML 属性、shell 引号、JSON 字符串包装之间没有互相打断
-- [ ] 结构合法：`<slide>` 下只放 `<style>`、`<data>`、`<note>`，文本都在 `<content>` 内
-- [ ] 路径正确：`<img src="@...">` 只在 `+create --slides` 的支持链路中使用
-
-高频失败信号和处理顺序：
-
-1. `invalid param` / 某一页创建失败
-2. 先检查失败页是否含未转义 `&` / `<` / `>`：`Q&A -> Q&amp;A`，属性 URL `a=1&b=2 -> a=1&amp;b=2`
-3. 再检查标签闭合、属性引号、`<content>` 结构
-4. 如果是 `--slides '[...]'`，怀疑 shell 截断时直接切两步创建法
-5. 创建后无论成功失败，都优先记录 `xml_presentation_id` 并回读确认是否已有部分页面写入
-
-### jq 命令模板（编辑已有 PPT 时使用）
-
-新建 PPT 推荐用 `+create --slides`。以下 jq 模板适用于向已有演示文稿追加页面的场景，可以避免手动转义双引号：
-
-```bash
-# 追加到末尾
-lark-cli slides xml_presentation.slide create \
-  --as user \
+lark-cli slides xml_presentation.slide create --as user \
   --params '{"xml_presentation_id":"YOUR_ID"}' \
-  --data "$(jq -n --arg content '<slide xmlns="http://www.larkoffice.com/sml/2.0">
-  <style><fill><fillColor color="BACKGROUND_COLOR"/></fill></style>
-  <data>
-    在这里放置 shape、line、table、chart 等元素
-  </data>
-</slide>' '{slide:{content:$content}}')"
-
-# 插到指定页之前：before_slide_id 必须在 --data body 里，与 slide 同级
-# ⚠️ 不要把 before_slide_id 写进 --params —— CLI 会当未知 query 参数静默下发，服务端忽略，新页跑到末尾
-lark-cli slides xml_presentation.slide create \
-  --as user \
-  --params '{"xml_presentation_id":"YOUR_ID"}' \
-  --data "$(jq -n --arg content '<slide ...>...</slide>' --arg before 'TARGET_SLIDE_ID' \
-    '{slide:{content:$content}, before_slide_id:$before}')"
+  --data "$(jq -n --arg content '<slide xmlns="http://www.larkoffice.com/sml/2.0"><data/></slide>' '{slide:{content:$content}}')"
 ```
 
-### 风格快速判断表
+To insert before an existing page, put `before_slide_id` in `--data`, not in `--params`.
 
-> **注意**：渐变色必须使用 `rgba()` 格式并带百分比停靠点，如 `linear-gradient(135deg,rgba(15,23,42,1) 0%,rgba(56,97,140,1) 100%)`。使用 `rgb()` 或省略停靠点会导致服务端回退为白色。
+## Media Upload
 
-| 场景/主题 | 推荐风格 | 背景 | 主色 | 文字色 |
-|----------|---------|------|------|-------|
-| 科技/AI/产品 | 深色科技风 | 深蓝渐变 `linear-gradient(135deg,rgba(15,23,42,1) 0%,rgba(56,97,140,1) 100%)` | 蓝色系 `rgb(59,130,246)` | 白色 |
-| 商务汇报/季度总结 | 浅色商务风 | 浅灰 `rgb(248,250,252)` | 深蓝 `rgb(30,60,114)` | 深灰 `rgb(30,41,59)` |
-| 教育/培训 | 清新明亮风 | 白色 `rgb(255,255,255)` | 绿色系 `rgb(34,197,94)` | 深灰 `rgb(51,65,85)` |
-| 创意/设计 | 渐变活力风 | 紫粉渐变 `linear-gradient(135deg,rgba(88,28,135,1) 0%,rgba(190,24,93,1) 100%)` | 粉紫色系 | 白色 |
-| 周报/日常汇报 | 简约专业风 | 浅灰 `rgb(248,250,252)` + 顶部彩色渐变条 | 蓝色 `rgb(59,130,246)` | 深色 `rgb(15,23,42)` |
-| 用户未指定 | 默认简约专业风 | 同上 | 同上 | 同上 |
+Slides XML image `src` must be a Lark `file_token`; do not use external HTTP(S) URLs.
 
-### 页面布局建议
+- New deck with `+create --slides`: `src="@./local.png"` is allowed and the shortcut uploads it.
+- Existing deck or raw `slide.create`: run `slides +media-upload` first, then write `src="<file_token>"`.
+- Existing slide edit: upload first, then use `+replace-slide` with `block_insert` or `block_replace`.
 
-| 页面类型 | 布局要点 |
-|---------|---------|
-| 封面页 | 居中大标题 + 副标题 + 底部信息，背景用渐变或深色 |
-| 数据概览页 | 指标卡片横排（rect 背景 + 大号数字 + 小号说明），下方列表或图表 |
-| 内容页 | 左侧竖线装饰 + 标题，下方分栏或列表 |
-| 对比/表格页 | table 元素或并列卡片，表头深色背景白字 |
-| 图表页 | chart 元素（column/line/pie），配合文字说明 |
-| 结尾页 | 居中感谢语 + 装饰线，风格与封面呼应 |
+Local paths must be safe paths under the current working directory. The upload limit is 20 MB.
 
-### 大纲模板
+## XML Rules
 
-生成大纲时使用以下格式，交给用户确认：
+- `<slide>` direct children are only `<style>`, `<data>`, and `<note>`.
+- Text belongs inside `<content><p>...</p></content>`.
+- Escape raw text before writing XML: `&` becomes `&amp;`, text `<` becomes `&lt;`, and text `>` becomes `&gt;`.
+- Gradient fills require `rgba()` stops with percentages, for example `linear-gradient(135deg,rgba(15,23,42,1) 0%,rgba(56,97,140,1) 100%)`.
+- For `xml_presentation.slide.replace`, `block_replace` needs the target block id and text shapes need `<content/>`; `+replace-slide` injects the required wrapper details.
 
-```text
-[PPT 标题] — [定位描述]，面向 [目标受众]
+## Validation
 
-模板：[未使用模板 / <category>/<template>.xml（推荐原因）]
+This execution skill validates at the XML/API layer. Before execution, check XML well-formedness, escaping, request body shape, and `lark-cli schema` output. Visual layout quality checks belong to creator workflows, not this execution layer.
 
-页面结构（N 页）：
-1. 封面页：[标题文案]
-2. [页面主题]：[要点1]、[要点2]、[要点3]
-3. [页面主题]：[要点描述]
-...
-N. 结尾页：[结尾文案]
+## Troubleshooting
 
-风格：[配色方案]，[排版风格]
-```
+| Symptom | Likely Cause | Next Action |
+| --- | --- | --- |
+| `400` XML or wrapper error | Bad XML or wrong `--data` shape | Check escaping, tag closure, and `lark-cli schema`. |
+| `403` permission denied | Wrong identity or missing scope | Confirm `--as user` vs `--as bot`; re-run auth for slides scope. |
+| `404` presentation/slide not found | Wrong token or unresolved wiki URL | Resolve wiki token or re-read current presentation. |
+| `1061002` media params error | Raw upload API used incorrectly | Use `slides +media-upload`; slides parent type is `slide_file`. |
+| `1061004` forbidden | Current identity cannot edit target deck | Use the owner identity or share the deck with the bot/user. |
+| `3350001` catch-all validation failure | XML not well-formed, bad replace wrapper, missing `<content/>`, or unescaped text | Run lint, inspect failed page XML, and prefer `+replace-slide` for block edits. |
+| `3350002` stale revision | `revision_id` is newer than current | Use `-1` or re-read the presentation and retry. |
+| Created deck has blank/missing pages | Shell/JSON argument truncation or escaping issue | Read back XML, then continue with two-step `slide.create`. |
+| Image does not show | `src` is URL or unresolved `@path` | Upload and replace with a `file_token`. |
 
-### 常用 Slide XML 模板
+## References
 
-可直接复制使用的模板（封面页、内容页、数据卡片页、结尾页）：[slide-templates.md](references/slide-templates.md)
-
----
-
-## 核心概念
-
-### URL 格式与 Token
-
-| URL 格式 | 示例 | Token 类型 | 处理方式 |
-|----------|------|-----------|----------|
-| `/slides/` | `https://example.larkoffice.com/slides/xxxxxxxxxxxxx` | `xml_presentation_id` | URL 路径中的 token 直接作为 `xml_presentation_id` 使用 |
-| `/wiki/` | `https://example.larkoffice.com/wiki/wikcnxxxxxxxxx` | `wiki_token` | ⚠️ **不能直接使用**，需要先查询获取真实的 `obj_token` |
-
-> `+replace-slide` 和 `+media-upload` shortcut 会自动解析以上两种 URL；直接调用原生 API 时仍需手动解析 wiki 链接。
-
-### Wiki 链接特殊处理（关键！）
-
-知识库链接（`/wiki/TOKEN`）背后可能是云文档、电子表格、幻灯片等不同类型的文档。**不能直接假设 URL 中的 token 就是 `xml_presentation_id`**，必须先查询实际类型和真实 token。
-
-#### 处理流程
-
-1. **使用 `wiki.spaces.get_node` 查询节点信息**
-   ```bash
-   lark-cli wiki spaces get_node --as user --params '{"token":"wiki_token"}'
-   ```
-
-2. **从返回结果中提取关键信息**
-   - `node.obj_type`：文档类型，幻灯片对应 `slides`
-   - `node.obj_token`：**真实的演示文稿 token**（用于后续操作）
-   - `node.title`：文档标题
-
-3. **确认 `obj_type` 为 `slides` 后，使用 `obj_token` 作为 `xml_presentation_id`**
-
-#### 查询示例
-
-```bash
-# 查询 wiki 节点
-lark-cli wiki spaces get_node --as user --params '{"token":"wikcnxxxxxxxxx"}'
-```
-
-返回结果示例：
-```json
-{
-   "node": {
-      "obj_type": "slides",
-      "obj_token": "xxxxxxxxxxxx",
-      "title": "2026 产品年度总结",
-      "node_type": "origin",
-      "space_id": "1234567890"
-   }
-}
-```
-
-```bash
-# 用 obj_token 读取幻灯片内容
-lark-cli slides xml_presentations get --as user --params '{"xml_presentation_id":"xxxxxxxxxxxx"}'
-```
-
-### 资源关系
-
-```text
-Wiki Space (知识空间)
-└── Wiki Node (知识库节点, obj_type: slides)
-    └── obj_token → xml_presentation_id
-
-Slides (演示文稿)
-├── xml_presentation_id (演示文稿唯一标识)
-├── revision_id (版本号)
-└── Slide (幻灯片页面)
-    └── slide_id (页面唯一标识)
-```
-
-## Shortcuts（推荐优先使用）
-
-Shortcut 是对常用操作的高级封装（`lark-cli slides +<verb> [flags]`）。有 Shortcut 的操作优先使用。
-
-| Shortcut | 说明 |
-|----------|------|
-| [`+create`](references/lark-slides-create.md) | 创建 PPT（可选 `--slides` 一步添加页面，支持 `<img src="@./local.png">` 占位符自动上传），bot 模式自动授权 |
-| [`+media-upload`](references/lark-slides-media-upload.md) | 上传本地图片到指定演示文稿，返回 `file_token`（用作 `<img src="...">`），最大 20 MB |
-| [`+replace-slide`](references/lark-slides-replace-slide.md) | 对已有幻灯片页面进行块级替换/插入（`block_replace` / `block_insert`），自动注入 id 和 `<content/>`，不改变页序 |
-
-## API Resources
-
-```bash
-lark-cli schema slides.<resource>.<method>   # 调用 API 前必须先查看参数结构
-lark-cli slides <resource> <method> [flags] # 调用 API
-```
-
-> **重要**：使用原生 API 时，必须先运行 `schema` 查看 `--data` / `--params` 参数结构，不要猜测字段格式。
-
-### xml_presentations
-
-  - `get` — 读取演示文稿全文信息，XML 格式返回
-
-### xml_presentation.slide
-
-  - `create` — 在指定 XML 演示文稿下创建页面
-  - `delete` — 在指定 XML 演示文稿下删除页面
-  - `get` — 获取指定 XML 演示文稿的单个页面 XML 内容
-  - `replace` — 对指定 XML 演示文稿页面进行元素级别的局部替换
-
-## 核心规则
-
-1. **先定模板/风格并出大纲再动手**：如果需求可匹配模板，先给用户 2-3 个模板候选；模板或自定义风格确定后，再生成大纲交给用户确认，避免返工
-2. **创建流程**：简单短 XML（1-3 页、结构简单、特殊字符少）可用 `slides +create --slides '[...]'` 一步创建；复杂内容、含图片/中文大段文本/嵌套引号/较多特殊字符，或超过 10 页时，默认先 `slides +create` 创建空白 PPT，再用 `xml_presentation.slide.create` 逐页添加
-3. **`<slide>` 直接子元素只有 `<style>`、`<data>`、`<note>`**：文本和图形必须放在 `<data>` 内
-4. **文本通过 `<content>` 表达**：必须用 `<content><p>...</p></content>`，不能把文字直接写在 shape 内
-5. **保存关键 ID**：后续操作需要 `xml_presentation_id`、`slide_id`、`revision_id`
-6. **删除谨慎**：删除操作不可逆，且至少保留一页幻灯片
-7. **编辑已有页面优先块级替换**：修改单个 shape/img 用 `+replace-slide`（`block_replace` / `block_insert`），不要整页重建；只有需要替换整页结构时才用 `slide.delete` + `slide.create`
-8. **`<img src>` 只能用上传到飞书 drive 的 `file_token`，禁止使用 http(s) 外链 URL**：飞书 slides 渲染端不会代理外链图片，外链 src 在 PPT 里通常不显示或显示破图。流程必须是「先把图存到本地 → 用 `slides +media-upload` 上传或 `+create --slides` 的 `@./path` 占位符自动上传 → 拿 `file_token` 写进 `<img src>`」。如果用户给了网图链接，先 `curl`/下载到 CWD 内再走上传流程，不要直接把外链 URL 塞进 `src`。**图片最大 20 MB**（slides upload API 不支持分片上传）。
-
-## 权限表
-
-| 方法 | 所需 scope |
-|------|-----------|
-| `slides +create` | `slides:presentation:create`, `slides:presentation:write_only`（含 `@` 占位符时还需 `docs:document.media:upload`） |
-| `slides +media-upload` | `docs:document.media:upload`（wiki URL 解析还需 `wiki:node:read`） |
-| `slides +replace-slide` | `slides:presentation:update`（wiki URL 解析还需 `wiki:node:read`） |
-| `xml_presentations.get` | `slides:presentation:read` |
-| `xml_presentation.slide.create` | `slides:presentation:update` 或 `slides:presentation:write_only` |
-| `xml_presentation.slide.delete` | `slides:presentation:update` 或 `slides:presentation:write_only` |
-| `xml_presentation.slide.get` | `slides:presentation:read` |
-| `xml_presentation.slide.replace` | `slides:presentation:update` |
-
-## 常见错误速查
-
-| 错误码 | 含义 | 解决方案 |
-|--------|------|----------|
-| 400 | XML 格式错误 | 检查 XML 语法，确保标签闭合 |
-| 400 | 请求包装错误 | 检查 `--data` 是否按 schema 传入 `xml_presentation.content` 或 `slide.content` |
-| 创建成功但页面空白/内容缺失/布局错乱 | 常见于 `--slides '[...]'` 的 shell 转义或长参数传递问题 | 改用两步创建：先 `slides +create`，再用 `jq -n` 包装 `xml_presentation.slide.create` 逐页添加，并在创建后立即读取 XML 验证 |
-| 404 | 演示文稿不存在 | 检查 `xml_presentation_id` 是否正确 |
-| 404 | 幻灯片不存在 | 检查 `slide_id` 是否正确 |
-| 403 | 权限不足 | 检查是否拥有对应的 scope |
-| 400 | 无法删除唯一幻灯片 | 演示文稿至少保留一页幻灯片 |
-| 1061002 | params error（媒体上传时） | 用 `slides +media-upload`，不要手拼原生 `medias/upload_all`；slides 唯一可用 `parent_type` 是 `slide_file` |
-| 1061004 | forbidden：当前身份对演示文稿无编辑权限 | 确认 user/bot 对目标 PPT 有编辑权限；bot 常见于 PPT 非该 bot 创建，需先授权或用 `+create --as bot` 新建 |
-| 3350001 | XML 非 well-formed、XML 结构不符合服务端要求，或 `xml_presentation.slide.replace` 失败（catch-all） | 优先检查未转义 `&` / `<` / `>`：`Q&A -> Q&amp;A`，属性 URL `a=1&b=2 -> a=1&amp;b=2`；运行 `layout_lint.py --input <file>` 定位行列和上下文；再检查 replace 场景的 `block_id` / `<content/>` / 坐标 |
-| 3350002 | `revision_id` 大于当前版本 | 用 `-1` 取当前版本，或重新读 `xml_presentations.get` 取最新 `revision_id` |
-| validation: unsafe file path | `--file` 给了绝对路径或上层路径 | `--file` 必须是 CWD 内相对路径；先 `cd` 到素材目录再执行 |
-
-## 创建前自查
-
-逐页生成 XML 前，快速检查：
-
-- [ ] 每页背景色/渐变是否设置？风格是否与整体一致？
-- [ ] 标题用大字号（28-48），正文用小字号（13-16），层级分明？
-- [ ] 同类元素配色一致？（如所有指标卡片同色系、所有正文同色）
-- [ ] 装饰元素（分割线、色块、竖线）颜色是否与主色协调？
-- [ ] 文本框尺寸是否足够容纳内容？（宽度 × 高度）
-- [ ] shape 的 `type` 是否正确？（文本框用 `text`，装饰用 `rect`）
-- [ ] XML 标签是否全部正确闭合？特殊字符（`&`、`<`、`>`）是否转义？
-
-## 症状 → 修复表
-
-| 看到的问题 | 改什么 |
-|-----------|--------|
-| 文字被截断/看不全 | 增大 shape 的 `width` 或 `height` |
-| 元素重叠 | 调整 `topLeftX`/`topLeftY`，拉开间距 |
-| 页面大面积空白 | 缩小元素间距，或增加内容填充 |
-| 文字和背景色太接近 | 深色背景用浅色文字，浅色背景用深色文字 |
-| 表格列宽不合理 | 调整 `colgroup` 中 `col` 的 `width` 值 |
-| 图表没有显示 | 检查 `chartPlotArea` 和 `chartData` 是否都包含，`dim1`/`dim2` 数据数量是否匹配 |
-| 图片被裁掉一部分 | `<img>` 的 `width`/`height` 是裁剪后尺寸，比例和原图不一致时会自动裁剪；要整图显示就让 `width:height` 对齐原图比例 |
-| 只想改某页的单个元素（文字/图片/形状） | 用 `+replace-slide` 块级替换，不要整页重建 |
-| 想给已有页加一张图（不动原有元素） | ① `+media-upload` 拿 `file_token` ② `+replace-slide` 用 `block_insert` 插入 `<img src="<file_token>" .../>`；不要再用 "整页 create + delete" 的老流程 |
-| 新插入的 `<img>` 挡住/重叠原有元素 | `slide.get` 读原页，对照已有块的 `topLeftX/Y/width/height` 挑空白位置；空间不够就在同一批 `--parts` 里先 `block_replace` 缩小/挪动现有块再 `block_insert` 图片 |
-| 渐变背景变成白色 | 渐变必须用 `rgba()` 格式 + 百分比停靠点，如 `linear-gradient(135deg,rgba(30,60,114,1) 0%,rgba(59,130,246,1) 100%)`；用 `rgb()` 或省略停靠点会被回退为白色 |
-| 渐变方向不对 | 调整 `linear-gradient` 的角度（`90deg` 水平、`180deg` 垂直、`135deg` 对角线） |
-| 整体风格不统一 | 封面页和结尾页用同一背景，内容页保持一致的配色和字号体系 |
-| API 返回 400 | 检查 XML 语法：标签闭合、属性引号、特殊字符转义 |
-| API 返回 3350001 | `block_replace` 根元素缺 `id=<block_id>` 或 `<shape>` 缺 `<content/>`，详见 replace-slide 文档 |
-| 图片不显示 / `<img src>` 仍是 `@path` | `@` 占位符**只在 `+create --slides` 中替换**；直接调 `xml_presentation.slide.create` 必须先用 `+media-upload` 拿 `file_token` 写进 src |
-| 上传图片报 1061002 params error | `parent_type` 必须是 `slide_file`（slides 唯一接受值）；不要手拼，用 `slides +media-upload` |
-
-## 参考文档
-
-| 文档 | 说明 |
-|------|------|
-| [lark-slides-create.md](references/lark-slides-create.md) | **+create Shortcut：创建 PPT（支持 `--slides` 一步添加页面，含 `@` 占位符自动上传图片）** |
-| [lark-slides-media-upload.md](references/lark-slides-media-upload.md) | **+media-upload Shortcut：上传本地图片，返回 `file_token`** |
-| [lark-slides-replace-slide.md](references/lark-slides-replace-slide.md) | **+replace-slide Shortcut：块级替换/插入，含合法根元素速查与 3350001 排错** |
-| [lark-slides-edit-workflows.md](references/lark-slides-edit-workflows.md) | 编辑已有页面的读-改-写流程与 action 决策树 |
-| [template-index.json](references/template-index.json) | **脚本缓存/轻量路由索引：由 `template_tool.py search` 使用，不是默认阅读入口** |
-| [template-catalog.md](references/template-catalog.md) | **按场景/色调匹配现成 PPT 模板，并定位到页型范围** |
-| [`scripts/template_tool.py`](scripts/template_tool.py) | **可选 Python 辅助脚本：`search` / `summarize` / `extract`，支持 `--layout-tag` 与 `extract --with-summary`** |
-| [`scripts/layout_lint.py`](scripts/layout_lint.py) | **本地预检脚本：先检查 XML well-formed，再检测重叠、越界、页脚碰撞、文本高度风险；不是完整 XSD schema 校验** |
-| [xml-schema-quick-ref.md](references/xml-schema-quick-ref.md) | **XML Schema 精简速查（必读）** |
-| [slide-templates.md](references/slide-templates.md) | 可复制的 Slide XML 模板 |
-| [xml-format-guide.md](references/xml-format-guide.md) | XML 详细结构与示例 |
-| [examples.md](references/examples.md) | CLI 调用示例 |
-| [slides_demo.xml](references/slides_demo.xml) | 真实 PPT 的完整 XML |
-| [slides_xml_schema_definition.xml](references/slides_xml_schema_definition.xml) | **完整 Schema 定义**（唯一协议依据） |
-| [lark-slides-xml-presentations-get.md](references/lark-slides-xml-presentations-get.md) | 读取 PPT 命令详情 |
-| [lark-slides-xml-presentation-slide-create.md](references/lark-slides-xml-presentation-slide-create.md) | 添加幻灯片命令详情 |
-| [lark-slides-xml-presentation-slide-delete.md](references/lark-slides-xml-presentation-slide-delete.md) | 删除幻灯片命令详情 |
-| [lark-slides-xml-presentation-slide-get.md](references/lark-slides-xml-presentation-slide-get.md) | 读取单个幻灯片命令详情 |
-| [lark-slides-xml-presentation-slide-replace.md](references/lark-slides-xml-presentation-slide-replace.md) | 原生 slide.replace API 命令详情 |
-
-> **注意**：如果 md 内容与 `slides_xml_schema_definition.xml` 或 `lark-cli schema slides.<resource>.<method>` 输出不一致，以后两者为准。
+| Reference | Purpose |
+| --- | --- |
+| [xml-schema-quick-ref.md](references/xml-schema-quick-ref.md) | Required XML element and attribute quick reference. |
+| [xml-format-guide.md](references/xml-format-guide.md) | Detailed XML structure and examples. |
+| [slides_xml_schema_definition.xml](references/slides_xml_schema_definition.xml) | Full XML schema definition. |
+| [lark-slides-create.md](references/lark-slides-create.md) | `+create` shortcut. |
+| [lark-slides-media-upload.md](references/lark-slides-media-upload.md) | `+media-upload` shortcut. |
+| [lark-slides-replace-slide.md](references/lark-slides-replace-slide.md) | `+replace-slide` shortcut. |
+| [lark-slides-edit-workflows.md](references/lark-slides-edit-workflows.md) | Existing-slide read/modify/write workflows. |
+| [lark-slides-xml-presentations-get.md](references/lark-slides-xml-presentations-get.md) | Raw presentation read API. |
+| [lark-slides-xml-presentation-slide-create.md](references/lark-slides-xml-presentation-slide-create.md) | Raw slide create API. |
+| [lark-slides-xml-presentation-slide-delete.md](references/lark-slides-xml-presentation-slide-delete.md) | Raw slide delete API. |
+| [lark-slides-xml-presentation-slide-get.md](references/lark-slides-xml-presentation-slide-get.md) | Raw slide get API. |
+| [lark-slides-xml-presentation-slide-replace.md](references/lark-slides-xml-presentation-slide-replace.md) | Raw slide replace API. |
+| [examples.md](references/examples.md) | CLI examples. |
+| [slides_demo.xml](references/slides_demo.xml) | Example presentation XML. |

skills/lark-slides/references/lark-slides-xml-presentation-slide-create.md

@@ -178,7 +178,7 @@ lark-cli slides xml_presentation.slide create --as user \
 | 400 | XML 格式错误 | 检查 `slide.content` 是否是完整 `<slide>` 元素 |
 | 400 | 请求体结构错误 | 检查是否按 `slide.content` 和 `before_slide_id` 包装 |
 | 403 | 权限不足 | 检查是否拥有 `slides:presentation:update` 或 `slides:presentation:write_only` scope |
-| 3350001 | XML 非 well-formed 或服务端参数校验失败 | 优先检查未转义字符：文本 `Q&A -> Q&amp;A`，文本 `<` / `>` 写成 `&lt;` / `&gt;`，属性 URL `a=1&b=2 -> a=1&amp;b=2`；创建前运行 `python3 skills/lark-slides/scripts/layout_lint.py --input <file>` 获取行列和上下文 |
+| 3350001 | XML 非 well-formed 或服务端参数校验失败 | 优先检查未转义字符：文本 `Q&A -> Q&amp;A`，文本 `<` / `>` 写成 `&lt;` / `&gt;`，属性 URL `a=1&b=2 -> a=1&amp;b=2`；再检查标签闭合、命名空间、`<content>` 结构和请求体包装 |
 
 ## 注意事项
 
@@ -188,7 +188,7 @@ lark-cli slides xml_presentation.slide create --as user \
 4. **fill / border 写法**: 颜色填充使用 `<fill><fillColor color="..."/></fill>`，边框常用 `<border color="..." width="2"/>`
 5. **插入位置**: 通过 `before_slide_id` 指定插入目标，而不是用 `position`
 6. **JSON 转义**: 如果直接内联 XML，需要正确转义双引号
-7. **本地预检**: 创建前运行 `layout_lint.py --input <file>`；它检查 XML well-formed 和布局风险，不等价于完整 XSD schema 校验
+7. **执行层预检**: 检查 XML well-formed、特殊字符转义、`slide.content` 包装和 `before_slide_id` 位置；布局质量检查不属于本执行层
 8. **建议**: 先使用 `xml_presentations.get` 获取现有结构，再添加新页面
 
 ## 批量添加建议

skills/lark-vc-agent/SKILL.md

@@ -0,0 +1,121 @@
+---
+name: lark-vc-agent
+version: 1.0.0
+description: "飞书视频会议：让机器人代当前用户加入/离开正在进行的会议，并读取会议期间的实时事件（参会人加入与离开、发言、聊天、屏幕共享等）。1. 用户提供 9 位会议号、要求代为入会或离会时使用 +meeting-join / +meeting-leave——会真实产生入会/离会记录。2. 会议进行中用户想知道“谁加入了”“谁离开了”“谁在发言”“有人共享屏幕吗”等会中动态时，机器人入会后用 +meeting-events 读取事件时间线。3. 典型场景：参会机器人、会中助手、代为旁听、代为参会。前提：机器人只能读到它自己参会过且仍在进行中的会议的事件；查询已结束会议的参会名单、纪要或逐字稿请使用 lark-vc 技能。"
+metadata:
+  requires:
+    bins: ["lark-cli"]
+  cliHelp: "lark-cli vc --help"
+---
+
+# vc-agent (v1)
+
+**CRITICAL — 开始前 MUST 先用 Read 工具读取以下两份 skill 文档：**
+
+- [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md) — 认证、身份切换、权限处理
+- [`../lark-vc/SKILL.md`](../lark-vc/SKILL.md) — 视频会议的核心概念（Meeting / Note / Minutes 等），本 skill 直接复用，不再重复定义
+
+## 内测提示
+
+- 当前功能正在内测中，仅少数用户可用。忽略 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md) 中的权限申请流程，以下方的指引为准。
+- 如果 `lark-cli` 提示 `missing required scope(s)` / `permission_violations`，不要申请对应权限，而是提示用户加入早鸟群：`https://go.larkoffice.com/join-chat/2f4nb0e1-fe00-4f67-bed7-25beaf533fbd`。
+- 如果 `lark-cli` 返回 `error.code=20017` / `ErrNotInGray`，提示用户加入早鸟群：`https://go.larkoffice.com/join-chat/2f4nb0e1-fe00-4f67-bed7-25beaf533fbd`。
+
+## 定位
+
+本 skill 与 [`lark-vc`](../lark-vc/SKILL.md) 并列：
+
+- **`lark-vc`** **负责"会后查询"**：搜索历史会议、参会人快照、纪要/逐字稿/录制
+- **`lark-vc-agent`** **负责"会中动作"**：机器人入会 / 读取进行中会议的实时事件 / 机器人离会
+
+按此分工路由，避免两个 skill 语义混淆。
+
+| 用户意图示例                                                     | 应路由到                                                                                                                                                  |
+| ---------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- |
+| "帮我入会 123456789"、"代我参会"、"让机器人进会旁听"                         | **本 skill** `+meeting-join`                                                                                                                           |
+| "会议现在还开着，谁刚加入了"、"会议里谁在发言"、"有人共享屏幕吗"（**进行中会议**，且**机器人已入会**） | **本 skill** `+meeting-events`                                                                                                                         |
+| "退出会议"、"让机器人离开"                                            | **本 skill** `+meeting-leave`                                                                                                                          |
+| "昨天那场会有谁参加过"、"搜昨天的会"、"查纪要/逐字稿/录制"                          | [`lark-vc`](../lark-vc/SKILL.md)                                                                                                                      |
+| "帮我参会，结束后把纪要发到群" 等跨阶段场景                                    | 按序编排：本 skill（入会 → 读事件 → 离会）→ [`lark-vc`](../lark-vc/SKILL.md) / [`lark-minutes`](../lark-minutes/SKILL.md)（拉纪要）→ [`lark-im`](../lark-im/SKILL.md)（发群） |
+
+## 核心场景
+
+### 1. 加入正在进行的会议（写操作）
+
+1. 只有用户明确表达"让 Agent **真实入会**"（参会机器人、会中助手、代为旁听、代参会）时才用 `+meeting-join`。只是查数据不要入会。
+2. `+meeting-join --meeting-number` 只接受 **9 位纯数字**会议号，不是会议链接整串、也不是 `meeting_id`。
+3. 返回体中的 `meeting.id` **必须立刻记录**——后续 `+meeting-events` / `+meeting-leave` 都靠它，**不能用 9 位会议号替代**。
+4. 入会对所有参会人可见，执行前核实 9 位会议号来源，避免误入错会。
+5. 仅支持 `user` 身份，需提前 `lark-cli auth login`。
+6. 若入会失败，优先查看 `+meeting-join` reference 的错误排查段落，重点确认会议号、密码、会议状态、等候室 / 审批以及会议是否禁止当前身份加入。
+
+### 2. 感知会中事件（读操作）
+
+1. 用户要看"会议里正在发生什么"（参会人加入/离开、聊天、转写、屏幕共享）时，用 `+meeting-events`。
+2. 输入是 **`meeting_id`**（长数字 ID），不是 9 位会议号。
+3. Bot 必须**真实参会过**（先 `+meeting-join`），否则事件流通常不可见。具体的状态边界、结束后宽限窗口与错误码（如 `10005 / 20001 / 20002`）请查看 `+meeting-events` reference。
+4. **不能做会后复盘**，**不能替代参会人快照查询**。如果会议已结束：
+   - 想拿纪要文档或逐字稿文档 token：用 `lark-cli vc +notes --meeting-ids <meeting.id>`
+   - 想拿 AI 产物（summary / todos / chapters）或导出逐字稿文件：先用 `lark-cli vc +recording --meeting-ids <meeting.id>` 拿 `minute_token`，再用 `lark-cli vc +notes --minute-tokens <minute_token>`
+   - 想看参会人快照：用 `vc meeting get --with-participants`（见 [`lark-vc`](../lark-vc/SKILL.md)）
+5. **默认必须使用** **`--page-all`**，除非用户明确要求“只查一页”，或确实需要控制返回体大小。
+6. 输出格式默认优先 `--format pretty`（时间线更易读）；只有在需要完整保留原始消息流与结构化字段时，才使用 `--format json`。
+7. **必须识别分页信号**：只要响应里出现 `has_more=true`、pretty 里的 `more available`，或返回了非空 `page_token`，就不能把当前结果当作完整事件流；默认应继续分页，或明确告诉用户当前只是部分结果。
+8. 保留响应里的 `page_token`，下次增量拉取直接续，不要从头再拉。
+9. **只要你是基于** **`+meeting-events`** **来回答一场正在进行中的会议内容，就不能直接复用旧结果。** 无论用户是在问“现在/刚刚/最新”的状态，还是让你“总结一下这个会议讲什么”，都必须先重新拉一次当前事件流，确认拿到的是最新信息，再基于最新结果回答。只有在用户明确要求基于某次历史快照继续分析时，才可以复用旧结果。
+
+### 3. 离开会议（写操作）
+
+1. 任务完成、或用户要求结束时，用 `+meeting-leave --meeting-id <从 +meeting-join 拿到的 meeting.id>`。
+2. `--meeting-id` **必须**是 `+meeting-join` 返回的长数字 `meeting.id`，**不接受 9 位会议号**。
+3. 离会**立即生效**，机器人从会议的参会人列表中消失，对其他参会人可见；若需要重新入会，再跑一次 `+meeting-join` 即可（非真正"不可逆"）。
+4. 仅支持 `user` 身份。
+
+### 4. Agent 参会最小闭环示范
+
+```bash
+# 1. 入会，捕获 meeting.id
+JOIN=$(lark-cli vc +meeting-join --meeting-number 123456789 --format json)
+MID=$(echo "$JOIN" | jq -r '.data.meeting.id')
+
+# 2. 会中轮询事件
+#    默认用 --page-all 拉全当前可见事件；下次增量优先复用 page_token
+#    典型间隔 10-30 秒
+lark-cli vc +meeting-events --meeting-id "$MID" --page-all --format pretty
+
+# 3. 任务完成或用户要求结束时离会
+lark-cli vc +meeting-leave --meeting-id "$MID"
+
+# 4. 会后可选：取纪要 / 逐字稿（跨到 lark-vc）
+lark-cli vc +notes --meeting-ids "$MID"
+```
+
+## Shortcuts
+
+Shortcut 是对常用操作的高级封装（`lark-cli vc +<verb> [flags]`）。
+
+| Shortcut                                                        | 类型 | 说明                                                                         |
+| --------------------------------------------------------------- | -- | -------------------------------------------------------------------------- |
+| [`+meeting-join`](references/lark-vc-agent-meeting-join.md)     | 写  | Join an in-progress meeting by 9-digit meeting number                      |
+| [`+meeting-events`](references/lark-vc-agent-meeting-events.md) | 读  | List bot meeting events (participant joined/left, transcript, chat, share) |
+| [`+meeting-leave`](references/lark-vc-agent-meeting-leave.md)   | 写  | Leave a meeting by meeting\_id                                             |
+
+- 使用 `+meeting-join` 前**必须**阅读 [references/lark-vc-agent-meeting-join.md](references/lark-vc-agent-meeting-join.md)，了解入参格式与写操作可见性风险。
+- 使用 `+meeting-events` 前**必须**阅读 [references/lark-vc-agent-meeting-events.md](references/lark-vc-agent-meeting-events.md)，了解 `meeting_id` 来源、分页、错误码（10005 / 20001 / 20002）与 "bot 仍在会中" 硬约束。
+- 使用 `+meeting-leave` 前**必须**阅读 [references/lark-vc-agent-meeting-leave.md](references/lark-vc-agent-meeting-leave.md)，了解 `meeting_id` 的来源与写操作可见性。
+
+## 权限表
+
+| Shortcut          | 所需 scope                       |
+| ----------------- | ------------------------------ |
+| `+meeting-join`   | `vc:meeting.bot.join:write`    |
+| `+meeting-events` | `vc:meeting.meetingevent:read` |
+| `+meeting-leave`  | `vc:meeting.bot.join:write`    |
+
+## 延伸
+
+- 查已结束会议、参会人快照、搜索历史会议 → [`lark-vc`](../lark-vc/SKILL.md)
+- 会议纪要、逐字稿 → [`lark-vc`](../lark-vc/SKILL.md) 的 `+notes`
+- 妙记产物（AI 总结 / 转写 / 章节）→ [`lark-minutes`](../lark-minutes/SKILL.md)
+- 会后把产物发到群 / 私聊 → [`lark-im`](../lark-im/SKILL.md)
+- 认证、身份切换、scope 管理 → [`lark-shared`](../lark-shared/SKILL.md)

skills/lark-vc-agent/references/lark-vc-agent-meeting-events.md

@@ -0,0 +1,247 @@
+
+# vc +meeting-events
+
+> **前置条件：** 先阅读 [`../lark-shared/SKILL.md`](../../lark-shared/SKILL.md) 了解认证、全局参数和安全规则。
+
+查询当前 bot 在一场正在进行的视频会议中收到的会中事件列表。该命令是**读操作**。对进行中会议，要求 bot 当前仍在会中；对已结束会议，存在一个**结束后 5 分钟内的宽限窗口**，只要 bot 曾经在这场会里出现过，仍可继续拉取事件。
+
+本 skill 对应 shortcut：`lark-cli vc +meeting-events`（调用 `GET /open-apis/vc/v1/bots/events`）。
+
+## 命令
+
+```bash
+# 默认用法：全量拉取当前可见事件
+lark-cli vc +meeting-events --meeting-id 69xxxxxxxxxxxxx28 --page-all --format pretty
+
+# 指定时间范围，并拉全该时间窗内当前可见事件
+lark-cli vc +meeting-events --meeting-id 69xxxxxxxxxxxxx28 --start 2026-04-17T15:00:00+08:00 --end 2026-04-17T16:00:00+08:00 --page-all --format pretty
+
+# 基于上一次保存的 page_token 继续查新增事件
+lark-cli vc +meeting-events --meeting-id 69xxxxxxxxxxxxx28 --page-token <last_page_token> --page-all --format pretty
+
+# 调试或控制返回体大小时，显式只查一页
+lark-cli vc +meeting-events --meeting-id 69xxxxxxxxxxxxx28 --page-size 20 --format json
+
+# 预览 API 调用（不实际请求）
+lark-cli vc +meeting-events --meeting-id 69xxxxxxxxxxxxx28 --dry-run
+```
+
+## 参数
+
+| 参数 | 必填 | 说明 |
+|------|------|------|
+| `--meeting-id <id>` | 是 | 会议 ID（长数字 ID，不是 9 位会议号） |
+| `--start <time>` | 否 | 起始时间，支持 ISO 8601 / `YYYY-MM-DD` / Unix 秒 |
+| `--end <time>` | 否 | 结束时间，支持 ISO 8601 / `YYYY-MM-DD` / Unix 秒 |
+| `--page-token <token>` | 否 | 从指定分页游标继续拉取下一页 |
+| `--page-size <n>` | 否 | 单页模式每页大小。CLI 会自动夹紧到 `20-100`；传 `--page-all` 时固定使用 `100` |
+| `--page-all` | 否 | 自动分页，直到没有更多页面为止（内部有安全上限） |
+| `--format <fmt>` | 否 | 输出格式：json (CLI 默认) / pretty（本 skill 推荐默认） / table / ndjson / csv |
+| `--dry-run` | 否 | 预览 API 调用，不执行 |
+
+## 核心约束
+
+### 1. 输入必须是 meeting_id，不是 9 位会议号
+
+`--meeting-id` 必须是会议的长数字 ID。它通常来自：
+- `+meeting-join` 返回体中的 `meeting.id`
+- `+search` 结果中的 `id`
+
+**不要**把 9 位会议号（`--meeting-number`）传给这个命令。
+
+### 2. 仅支持 user 身份
+
+该命令仅支持 `user` 身份。
+
+### 3. bot 必须在会中，或在会议结束后的 5 分钟宽限窗口内曾经在会中
+
+这是查询“bot 在会中观察到的事件”的接口。若 bot 已离会、未入会、或会议已经无法再判断 bot 身份，后端通常会报：
+- `bot is not in meeting, no permission`
+
+因此，最稳妥的调用顺序通常是：
+
+```bash
+# 先入会
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 记录返回的 meeting.id
+
+# 再查询事件
+lark-cli vc +meeting-events --meeting-id <meeting.id>
+```
+
+更精确地说，后端当前的判断规则是：
+
+- **会议进行中**：要求 bot **当前仍在会中**
+- **会议已结束后的 5 分钟内**：只要 bot **曾经在这场会中出现过**，仍可拉取事件
+- **会议结束超过 5 分钟**：按会议结束处理，通常不再返回事件流
+- **bot 从未真实入会过**：即使会议仍在进行或刚结束，也会返回 `10005 bot is not in meeting`
+
+### 4. 自动分页规则
+
+- **先分清两层默认值**：
+  - shortcut 本身：不传 `--page-all` 时，只查 1 页。
+  - 本 skill 的默认策略：除非用户明确要求只看一页，或你确实需要控制返回体大小，否则默认**必须主动带 `--page-all`**，把当前可见事件尽量一次拉全。
+- 传 `--page-all`：开启自动分页，直到没有更多页面为止。
+- `--page-all` 时，CLI 固定使用最大 `page_size=100`。
+
+执行准则：
+
+- **默认命令模板**：`lark-cli vc +meeting-events --meeting-id <meeting.id> --page-all --format pretty`
+- 如果你发现自己执行成了不带 `--page-all` 的单页查询，而响应里又出现 `has_more=true` / `more available` / 非空 `page_token`，应立刻意识到这只是部分结果。
+- 遇到上述情况，默认补救方式是继续使用返回的 `page_token` 续拉，例如：`lark-cli vc +meeting-events --meeting-id <meeting.id> --page-token <returned_page_token> --page-all --format pretty`
+- 只有在用户明确要求“就看第一页”“先不要翻页”时，才不要默认带 `--page-all`
+- 只要你是基于 `+meeting-events` 来回答一场**正在进行中的会议内容**，就不能直接复用上一次查询结果。无论用户是在问“现在是谁在说话”“刚刚发生了什么”“最新事件有哪些”，还是让你“总结一下这个会议讲什么”，都必须先重新执行一次 `+meeting-events`，确认拿到的是最新事件流，再回答用户。只有在用户明确要求基于某次历史快照继续分析时，才可以复用旧结果。
+
+### 5. pretty / json 输出差异
+
+- `--format pretty`：输出会议主题、会议时间和逐条时间线，适合快速理解“发生了什么”，也是本 skill 的默认推荐格式。
+- `--format json`：保留完整原始 `events[]` 结构——参会人 open_id、聊天原文、share_doc、分页字段都在原始响应里，适合提取字段、联动其他命令或做进一步程序处理。
+
+**选型原则**：只要目标是告诉用户“发生了什么”，默认就用 `--page-all --format pretty`；只有在需要完整原始消息流和结构化字段时，才改用 `json`。
+
+> **注意**：pretty 输出中的正文文本会做单行转义，真实换行会显示为 `\n`，避免打乱时间线布局。
+
+### 6. 内容理解模式：共享文档不能只看标题
+
+当用户意图是：
+
+- “总结这个会议”
+- “这个会议讲了什么”
+- “有哪些结论 / 待办 / 关键讨论”
+- “共享文档里在讲什么”
+
+不要只基于事件时间线直接回答。此时 `+meeting-events` 只是**线索发现器**，不是最终信息源。
+
+执行准则：
+
+- 这类问题默认先用 `lark-cli vc +meeting-events --meeting-id <meeting.id> --page-all --format json` 拉取最新事件流。
+- 如果事件中出现共享文档线索，例如：
+  - `magic_share_started`
+  - `share_doc.title`
+  - `share_doc.url`
+- 必须继续读取共享文档内容，再生成总结，不能只根据“开始共享了某文档”这条事件和文档标题来概括会议内容。
+- 若存在多个共享文档，优先读取**最近一次共享**的文档。
+- 若文档读取失败，必须明确说明“以下总结仅基于会中事件流，未成功读取共享文档内容”。
+
+### 7. 关于 `page_token` 的返回与续拉
+
+- 不管这次是只查 1 页，还是通过 `--page-all` 已经把当前可见事件都拿完，都应把最后拿到的 `page_token` 一并保留下来并返回给用户。
+- 只要响应里出现 `has_more=true`、pretty 里出现 `more available`，或返回了非空 `page_token`，就必须先判断当前结果是否完整；默认情况下，这意味着你还需要继续分页。
+- 如果没有使用 `--page-all`，但出现了上述分页信号，默认应继续用返回的 `page_token` 拉下一页，而不是直接结束。只有在用户明确不要继续翻页时，才可以停止并明确说明当前结果不完整。
+- 下次继续“查新增事件”时，应优先复用上一次保存的 `page_token`，而不是从头全量再拉一次。
+- 只有在用户明确要求“从头回放全部事件”时，才忽略历史 `page_token`，重新从第一页开始。
+- 但如果用户要你回答的是**当前这场会正在讲什么**，而不是“上一次之后新增了什么”，也要先做一次新的事件查询，再决定是否需要基于旧 `page_token` 继续补拉。
+
+## 返回结构
+
+常见顶层字段：
+
+| 字段 | 说明 |
+|------|------|
+| `events` | 事件列表 |
+| `has_more` | 是否还有下一页 |
+| `page_token` | 下一页游标 |
+
+事件 `event_type` 常见类型：
+
+| event_type | 含义 |
+|-----------|------|
+| `participant_joined` | 有参会人加入会议 |
+| `participant_left` | 有参会人离开会议 |
+| `chat_received` | 收到会中聊天消息 |
+| `transcript_received` | 收到转写文本 |
+| `magic_share_started` | 开始共享内容 / 文档 |
+| `magic_share_ended` | 结束共享 |
+
+## pretty 输出示例
+
+```text
+会议主题：张三的视频会议
+会议时间：2026-04-17 15:28:52（进行中）
+
+[00:00:33] 明日之虾BOE(ou_xxx) 加入了会议
+[00:00:41] 张三(ou_xxx): [text] 6666
+[00:00:44] 张三(ou_xxx) 开始共享《智能纪要：飞书20251022-140223 2026年3月9日》
+           URL: https://...
+[00:01:32] 张三(ou_xxx): [reaction] JIAYI
+```
+
+## 如何获取输入参数
+
+| 输入参数 | 获取方式 |
+|---------|---------|
+| `meeting-id` | `+meeting-join` 返回的 `meeting.id`；或 `+search` 结果中的 `id` |
+| `start` / `end` | 用户给出的时间范围；如未给出则默认取全量可见事件 |
+| `page-token` | 上一页或上一次查询结果中保存的 `page_token`；建议持久化保存，便于下次继续拉取新增事件 |
+
+## Agent 组合场景
+
+### 场景 1：入会后查看会中发生了什么
+
+```bash
+# 第 1 步：加入会议，记录返回的 meeting.id
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 第 2 步：查询事件流
+lark-cli vc +meeting-events --meeting-id <meeting.id> --page-all --format pretty
+```
+
+### 场景 2：过滤某段时间内的事件
+
+```bash
+lark-cli vc +meeting-events \
+  --meeting-id <meeting.id> \
+  --start 2026-04-17T15:00:00+08:00 \
+  --end 2026-04-17T16:00:00+08:00 \
+  --page-all \
+  --format pretty
+```
+
+### 场景 3：基于上一次的 `page_token` 继续查新增事件
+
+```bash
+# 上一次查询结束后，保留最后返回的 page_token
+# 这次直接从该游标继续拉新增事件
+lark-cli vc +meeting-events \
+  --meeting-id <meeting.id> \
+  --page-token <last_page_token> \
+  --page-all \
+  --format pretty
+```
+
+适用规则：
+
+- 当用户说“继续看新事件”“看上次之后新增了什么”时，优先使用上一次保存的 `page_token`。
+- 如果这次返回里仍有 `has_more=true`、pretty 里出现 `more available`，或又返回了新的 `page_token`，说明新增事件还没拉完，应继续分页，而不是把当前页误当成完整增量结果。
+- 只有在用户明确要求“从头回放全部事件”时，才忽略已有 `page_token`，重新从第一页开始。
+
+## 常见错误与排查
+
+| 错误现象 | 根本原因 | 解决方案 |
+|---------|---------|---------|
+| `--meeting-id is required` | 未传入 `--meeting-id` | 传入长数字 `meeting.id` |
+| `10005 bot is not in meeting` | bot 从未真实入会该会议；或会议已结束但 bot 从未在会中出现过 | 先 `+meeting-join --meeting-number <9位号>` 真实入会再查；如果会议已经结束且当时 bot 没进过会，本接口也拉不到数据。**如果只是想看参会人快照，改用 `lark-cli vc meeting get --params '{"meeting_id":"<meeting.id>"}' --with-participants`**（不依赖 bot 身份参会） |
+| `20001 meeting_status_MEETING_END` | 会议已结束且已超出后端允许的 5 分钟宽限窗口 | 本接口不再适合继续拉取事件。若要拿纪要文档或逐字稿 token，用 `lark-cli vc +notes --meeting-ids <meeting.id>`；若要拿 AI 产物（summary / todos / chapters）或导出逐字稿文件，先用 `lark-cli vc +recording --meeting-ids <meeting.id>` 拿 `minute_token`，再用 `lark-cli vc +notes --minute-tokens <minute_token>`；参会人请用 `lark-cli vc meeting get --params '{"meeting_id":"<meeting.id>"}' --with-participants` |
+| `20002 meeting not exist` | `meeting_id` 错误，或会议实例当前已不可获取（常见于把 9 位会议号当 meeting_id 传） | 确认传入的是长数字 `meeting_id`，不是 9 位会议号 |
+| `HTTP 404` / `HTTP 500` | 服务端当前无法找到或处理该会议实例 | 换一个正在进行且 bot 可见的 meeting_id，或排查后端问题 |
+
+## 提示
+
+- 这是**会中事件流**查询，不适合拿来搜历史会议记录；搜历史会议请用 `+search`。
+- 如果会议已经结束，不要卡在 `+meeting-events`：  
+  - 想拿纪要文档或逐字稿 token：用 `lark-cli vc +notes --meeting-ids <meeting.id>`
+  - 想拿 AI 产物（summary / todos / chapters）或导出逐字稿文件：先用 `lark-cli vc +recording --meeting-ids <meeting.id>` 拿 `minute_token`，再用 `lark-cli vc +notes --minute-tokens <minute_token>`
+- 事件列表是否完整，取决于 bot 何时入会、何时离会，以及后端当前可见的会中事件范围。对于已结束会议，通常只在**结束后 5 分钟内**、且 bot **曾经在会中**时还能继续拉到事件。
+- 查询"谁参加过某会议"请用 `vc meeting get --params '{"meeting_id":"<id>","with_participants":true}'`——这是参会人**快照** API，不依赖 bot 是否参会，对已结束会议也可查；**不要** 用 `+meeting-events` 做参会人查询。
+
+## 参考
+
+- [lark-vc-agent-meeting-join](lark-vc-agent-meeting-join.md) — 先真实入会
+- [lark-vc-agent-meeting-leave](lark-vc-agent-meeting-leave.md) — 完成任务后离会
+- [lark-vc-search](../../lark-vc/references/lark-vc-search.md) — 搜索历史会议（获取 meeting_id）
+- [lark-vc-recording](../../lark-vc/references/lark-vc-recording.md) — 查询 minute_token
+- [lark-vc-notes](../../lark-vc/references/lark-vc-notes.md) — 获取会议纪要
+- [lark-vc-agent](../SKILL.md) — Agent 参会能力（本 skill）
+- [lark-vc](../../lark-vc/SKILL.md) — 视频会议原子域（Meeting / Note 等核心概念）
+- [lark-shared](../../lark-shared/SKILL.md) — 认证和全局参数

skills/lark-vc-agent/references/lark-vc-agent-meeting-join.md

@@ -0,0 +1,133 @@
+
+# vc +meeting-join
+
+> **前置条件：** 先阅读 [`../lark-shared/SKILL.md`](../../lark-shared/SKILL.md) 了解认证、全局参数和安全规则。
+
+通过 9 位会议号加入一场正在进行的视频会议（bot join）。这是一次**写操作**，会实际让当前身份加入会议。
+
+本 skill 对应 shortcut：`lark-cli vc +meeting-join`（调用 `POST /open-apis/vc/v1/bots/join`）。
+
+## 命令
+
+```bash
+# 仅指定会议号（无密码）
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 指定会议号 + 密码
+lark-cli vc +meeting-join --meeting-number 123456789 --password 8888
+
+# 输出格式
+lark-cli vc +meeting-join --meeting-number 123456789 --format json
+
+# 预览 API 调用（不实际加入会议）
+lark-cli vc +meeting-join --meeting-number 123456789 --dry-run
+```
+
+## 参数
+
+| 参数 | 必填 | 说明 |
+|------|------|------|
+| `--meeting-number <no>` | 是 | 会议号，必须为 **9 位纯数字** |
+| `--password <pw>` | 否 | 会议密码，仅在该会议设置了入会密码时传入 |
+| `--format <fmt>` | 否 | 输出格式：json (默认) / pretty / table / ndjson / csv |
+| `--dry-run` | 否 | 预览 API 调用，不执行 |
+
+## 核心约束
+
+### 1. 仅支持 user 身份
+
+该命令仅支持 `user` 身份。
+
+### 2. 会议号格式严格校验
+
+`--meeting-number` 必须是 9 位纯数字，否则本地校验直接报错：
+`--meeting-number must be exactly 9 digits`。
+
+常见错误来源：
+- 把会议链接整条粘进来（应仅取尾部的 9 位数字）
+- 把 `meeting_id`（长数字 ID）当成会议号传入（两者不是同一个东西）
+
+### 3. 会议必须已开始且允许入会
+
+- 会议必须处于**进行中**状态，bot 无法加入尚未开始或已结束的会议。
+- 若会议设置了**等候室 / 入会审批**，bot 可能需要主持人放行后才真正入会。
+- 若返回 `HTTP 403: no permission`（错误码 `121003`），不要只理解成“账号没权限”。这类报错更常见的原因是：会议参数或会控配置当前不满足入会条件，例如会议号填错、密码未传或错误、会议尚未开始、等候室 / 入会审批未放行、会议禁止外部/特定身份加入等。应先确认这些配置项，再重试。
+
+### 4. 机器人入会后对其他参会人可见
+
+这是一次真实入会操作，机器人会立即出现在参会人列表中，其他参会人可见，并产生会议日志。误入错会的社交成本高于技术成本——执行前优先确认 9 位会议号的来源（用户输入 / 会议链接末尾），不要臆造。参数格式有疑问时可用 `--dry-run` 预览请求体。
+
+## 输出结果
+
+接口返回会议基本信息，字段视具体响应而定，常见字段：
+
+| 字段 | 说明 |
+|------|------|
+| `meeting.id` | 会议 ID（可后续传给 `+meeting-leave --meeting-id`） |
+| `meeting.meeting_no` | 会议号（与入参一致） |
+| `meeting.topic` | 会议主题 |
+| `meeting.start_time` | 会议开始时间 |
+
+> **重要**：拿到 `meeting.id` 后务必保留，退出会议（`+meeting-leave`）需要使用它，而不是会议号。
+
+## 如何获取输入参数
+
+| 输入参数 | 获取方式 |
+|---------|---------|
+| `meeting-number` | 会议号由主持人分享；也可从会议链接尾部解析 9 位数字 |
+| `password` | 若会议设置了入会密码，由主持人提供 |
+
+## Agent 组合场景
+
+### 场景 1：加入会议 → 离开会议（最小闭环）
+
+```bash
+# 第 1 步：加入会议，记录返回的 meeting.id
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 第 2 步：完成任务后，使用上一步返回的 meeting.id 离开会议
+lark-cli vc +meeting-leave --meeting-id <meeting.id>
+```
+
+### 场景 2：加入会议 → 会后拉取纪要 / 录制
+
+```bash
+# 第 1 步：加入并参会
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 第 2 步：离会
+lark-cli vc +meeting-leave --meeting-id <meeting.id>
+
+# 第 3 步：会议结束后，查询录制（拿到 minute_token）
+lark-cli vc +recording --meeting-ids <meeting.id>
+
+# 第 4 步：查询会议纪要（总结 / 待办 / 章节 / 逐字稿）
+lark-cli vc +notes --meeting-ids <meeting.id>
+```
+
+## 常见错误与排查
+
+| 错误现象 | 根本原因 | 解决方案 |
+|---------|---------|---------|
+| `--meeting-number must be exactly 9 digits` | 会议号不是 9 位纯数字 | 检查是否误传了会议链接或 meeting_id |
+| 会议密码错误 | `--password` 错误或未提供 | 向主持人确认会议密码 |
+| 会议不存在 / 已结束 | 会议号错误或会议未进行中 | 确认会议正在进行中 |
+| `HTTP 403: no permission` / `121003` | 入会前置条件不满足，通常不是单纯 scope 问题 | 依次确认：1）会议允许智能体加入；2）会议号正确；3）如有密码，已正确传入 `--password`；4）会议已开始；5）等候室 / 入会审批已放行；6）会议未禁止当前身份加入（如限制外部、限制 bot、仅特定成员可入会）；确认后重试 |
+| 入会被拒绝 | 等候室 / 入会审批 / 限制外部入会 | 联系主持人放行或调整会议设置 |
+
+## 提示
+
+- 仅在 Agent 需要**真实加入**会议（例如参会机器人、会中助手）时使用；只拉取会议数据不需要入会。
+- 入会会让机器人立即出现在参会列表；若要回退，直接 `+meeting-leave` 即可。参数格式不确定时可选 `--dry-run` 预览，但不是必经步骤。
+- 执行成功后，立即记录返回的 `meeting.id`，用于后续 `+meeting-leave` / `+meeting-events`。
+
+## 参考
+
+- [lark-vc-agent-meeting-leave](lark-vc-agent-meeting-leave.md) — 对应的离会命令
+- [lark-vc-agent-meeting-events](lark-vc-agent-meeting-events.md) — 会中事件流
+- [lark-vc-search](../../lark-vc/references/lark-vc-search.md) — 搜索历史会议记录
+- [lark-vc-recording](../../lark-vc/references/lark-vc-recording.md) — 查询 minute_token
+- [lark-vc-notes](../../lark-vc/references/lark-vc-notes.md) — 获取会议纪要
+- [lark-vc-agent](../SKILL.md) — Agent 参会能力（本 skill）
+- [lark-vc](../../lark-vc/SKILL.md) — 视频会议原子域（Meeting / Note 等核心概念）
+- [lark-shared](../../lark-shared/SKILL.md) — 认证和全局参数

skills/lark-vc-agent/references/lark-vc-agent-meeting-leave.md

@@ -0,0 +1,111 @@
+
+# vc +meeting-leave
+
+> **前置条件：** 先阅读 [`../lark-shared/SKILL.md`](../../lark-shared/SKILL.md) 了解认证、全局参数和安全规则。
+
+通过 `meeting_id` 离开当前身份所在的视频会议（bot leave）。这是一次**写操作**，会实际把当前身份从会议中移出。
+
+本 skill 对应 shortcut：`lark-cli vc +meeting-leave`（调用 `POST /open-apis/vc/v1/bots/leave`）。
+
+## 命令
+
+```bash
+# 通过 meeting_id 离会
+lark-cli vc +meeting-leave --meeting-id 69xxxxxxxxxxxxx28
+
+# 输出格式
+lark-cli vc +meeting-leave --meeting-id 69xxxxxxxxxxxxx28 --format json
+
+# 预览 API 调用（不实际离会）
+lark-cli vc +meeting-leave --meeting-id 69xxxxxxxxxxxxx28 --dry-run
+```
+
+## 参数
+
+| 参数 | 必填 | 说明 |
+|------|------|------|
+| `--meeting-id <id>` | 是 | 会议 ID（**不是 9 位会议号**） |
+| `--format <fmt>` | 否 | 输出格式：json (默认) / pretty / table / ndjson / csv |
+| `--dry-run` | 否 | 预览 API 调用，不执行 |
+
+## 核心约束
+
+### 1. 入参是 meeting_id，不是会议号
+
+`--meeting-id` 必须是会议的长数字 ID，通常由 `+meeting-join` 返回体中的 `meeting.id` 提供，也可从 `+search` 结果中的 `id` 字段获取。**传 9 位会议号会失败**。
+
+### 2. 仅支持 user 身份
+
+该命令仅支持 `user` 身份。只能让当前身份自己离会，无法强制移出其他参会人。
+
+### 3. 当前身份必须在会议中
+
+必须先通过 `+meeting-join` 或其他方式在该会议中，否则接口会报错。
+
+### 4. 离会立即生效，对其他参会人可见
+
+机器人会立刻从参会列表消失；若会议启用了录制/纪要，bot 的参会时段到此截止。确认任务完成再调用；如需要重新入会，再跑 `+meeting-join` 即可（非真正"不可逆"）。
+
+## 输出结果
+
+接口成功返回时，默认输出：`Left meeting <meeting-id> successfully.`。
+`--format json` 返回 API 原始响应体。
+
+## 如何获取输入参数
+
+| 输入参数 | 获取方式 |
+|---------|---------|
+| `meeting-id` | `+meeting-join` 返回的 `meeting.id`；或 `+search` 结果中的 `id` 字段 |
+
+## Agent 组合场景
+
+### 场景 1：加入 → 完成任务 → 离开（最小闭环）
+
+```bash
+# 第 1 步：加入会议，记录 meeting.id
+lark-cli vc +meeting-join --meeting-number 123456789
+
+# 第 2 步：在会中完成任务（如监听发言、记录信息等）
+# ...
+
+# 第 3 步：使用上一步记录的 meeting.id 离会
+lark-cli vc +meeting-leave --meeting-id <meeting.id>
+```
+
+### 场景 2：会后补拉产物
+
+```bash
+# 第 1 步：离会后会议仍在进行或已结束
+lark-cli vc +meeting-leave --meeting-id <meeting.id>
+
+# 第 2 步：会议结束后查询录制
+lark-cli vc +recording --meeting-ids <meeting.id>
+
+# 第 3 步：查询会议纪要
+lark-cli vc +notes --meeting-ids <meeting.id>
+```
+
+## 常见错误与排查
+
+| 错误现象 | 根本原因 | 解决方案 |
+|---------|---------|---------|
+| `--meeting-id is required` | 未传入 `--meeting-id` | 传入从 `+meeting-join` 得到的 `meeting.id` |
+| `meeting not found` / `invalid meeting_id` | 误传了 9 位会议号 | 必须使用 `meeting.id`，不是会议号 |
+| `not in meeting` | 当前身份并不在该会议中 | 确认先 `+meeting-join` 成功 |
+
+## 提示
+
+- 离会会让机器人从参会列表消失，对其他参会人可见；若需要重新入会直接再 `+meeting-join`，不是真正的"不可逆"。参数格式不确定时可选 `--dry-run` 预览。
+- 与 `+meeting-join` 成对使用：能 join 的身份才能 leave。
+- `meeting_id` 必须来自 `+meeting-join` 的返回值，不要用 9 位会议号。
+
+## 参考
+
+- [lark-vc-agent-meeting-join](lark-vc-agent-meeting-join.md) — 对应的入会命令
+- [lark-vc-agent-meeting-events](lark-vc-agent-meeting-events.md) — 会中事件流
+- [lark-vc-search](../../lark-vc/references/lark-vc-search.md) — 搜索历史会议（获取 meeting_id）
+- [lark-vc-recording](../../lark-vc/references/lark-vc-recording.md) — 查询 minute_token
+- [lark-vc-notes](../../lark-vc/references/lark-vc-notes.md) — 获取会议纪要
+- [lark-vc-agent](../SKILL.md) — Agent 参会能力（本 skill）
+- [lark-vc](../../lark-vc/SKILL.md) — 视频会议原子域（Meeting / Note 等核心概念）
+- [lark-shared](../../lark-shared/SKILL.md) — 认证和全局参数

skills/lark-vc/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: lark-vc
 version: 1.0.0
-description: "飞书视频会议：查询会议记录、获取会议纪要产物（总结、待办、章节、逐字稿）。1. 查询已经结束的会议数量或详情时使用本技能(如历史日期｜ 昨天 | 上周 | 今天已经开过的会议等场景)，查询未开始的会议日程使用 lark-calendar 技能。2. 支持通过关键词、时间范围、组织者、参与者、会议室等筛选条件搜索会议记录。3. 获取或整理会议纪要时使用本技能。"
+description: "飞书视频会议：搜索历史会议、查询会议纪要产物（总结、待办、章节、逐字稿）、查询会议参会人快照。1. 查询已经结束的会议数量或详情时使用本技能（如历史日期｜昨天｜上周｜今天已经开过的会议等场景），查询未开始的会议日程使用 lark-calendar 技能。2. 支持通过关键词、时间范围、组织者、参与者、会议室等筛选条件搜索会议。3. 获取或整理会议纪要、逐字稿、录制产物时使用本技能。4. 查询“谁参加过某会议”“参会人列表”等参会人快照信息用 vc meeting get --with-participants（任意时点可查，含已结束会议）。注意：**Agent 真实入会/离会、感知正在进行中会议的实时事件**请使用 lark-vc-agent 技能，本技能不覆盖写操作和会中事件流。"
 metadata:
   requires:
     bins: ["lark-cli"]
@@ -14,8 +14,7 @@ metadata:
 
 ## 核心概念
 
-- **视频会议（Meeting）**：飞书视频会议实例，通过 meeting\_id 标识。
-- **会议记录（Meeting Record）**：视频会议结束后生成的记录，支持通过关键词、时间段、参会人、组织者、会议室等筛选条件搜索会议室。
+- **视频会议（Meeting）**：飞书视频会议实例，通过 meeting\_id 标识。已结束的会议支持通过关键词、时间段、参会人、组织者、会议室等条件搜索（见 `+search`）。
 - **会议纪要（Note）**：视频会议结束后生成的结构化文档，包含纪要文档（包含总结、待办、章节）和逐字稿文档。
 - **妙记（Minutes）**：来源于飞书视频会议的录制产物或用户上传的音视频文件，支持视频/音频的转写和会议纪要，通过 minute\_token 标识。
 - **纪要文档（MainDoc）**：AI 智能纪要的主文档，包含 AI 生成的总结和待办，对应 `note_doc_token`。
@@ -67,6 +66,23 @@ lark-cli drive metas batch_query --data '{"request_docs": [{"doc_type": "docx",
 lark-cli docs +fetch --api-version v2 --doc <doc_token> --doc-format markdown
 ```
 
+### 4. 查询参会人快照（读操作）
+
+用户问"谁参加过这场会议""这个会议有哪些参会人""某某参会了吗"等**参会人快照**类问题时，使用 **`vc meeting get --with-participants`**：这是参会人服务端快照 API，不依赖 bot 身份参会，**已结束会议也可查**：
+
+```bash
+lark-cli vc meeting get --params '{"meeting_id":"<meeting_id>","with_participants":true}'
+```
+
+选型判断表：
+
+| 用户意图 | 推荐命令 | 所在 skill |
+|---------|---------|--------|
+| 参会人快照（谁参加过、何时入/离会，任意时点）| `vc meeting get --with-participants` | 本 skill |
+| 已结束会议的发言内容 | `vc +notes` 取 `verbatim_doc_token` 再 `docs +fetch` | 本 skill |
+| **进行中会议**的实时事件流（转写、聊天、共享、会中加入/离开）| `vc +meeting-events` | [`lark-vc-agent`](../lark-vc-agent/SKILL.md) |
+| **Agent 真实入会 / 离会** | `vc +meeting-join` / `vc +meeting-leave` | [`lark-vc-agent`](../lark-vc-agent/SKILL.md) |
+
 ## 资源关系
 
 ```
@@ -109,6 +125,8 @@ Shortcut 是对常用操作的高级封装（`lark-cli vc +<verb> [flags]`）。
 - 使用 `+notes` 命令时，必须阅读 [references/lark-vc-notes.md](references/lark-vc-notes.md)，了解查询参数、产物类型和返回值结构。
 - 使用 `+recording` 命令时，必须阅读 [references/lark-vc-recording.md](references/lark-vc-recording.md)，了解查询参数和返回值结构。
 
+> **Agent 参会相关命令已独立**：`+meeting-join` / `+meeting-leave` / `+meeting-events` 请使用 [`lark-vc-agent`](../lark-vc-agent/SKILL.md) 技能。
+
 ## API Resources
 
 ```bash
@@ -146,3 +164,5 @@ lark-cli vc meeting get --params '{"meeting_id": "<meeting_id>", "with_participa
 | `+recording --calendar-event-ids` | `vc:record:readonly`、`calendar:calendar:read`、`calendar:calendar.event:read` |
 | `+search` | `vc:meeting.search:read` |
 | `meeting.get` | `vc:meeting.meetingevent:read` |
+
+> Agent 参会相关 scope（`vc:meeting.bot.join:write` / `vc:meeting.meetingevent:read`）见 [`lark-vc-agent`](../lark-vc-agent/SKILL.md)。

skills/lark-vc/references/lark-vc-recording.md

@@ -109,7 +109,7 @@ lark-cli vc +notes --minute-tokens <minute_token>
 
 ```bash
 # 第 1 步：搜索历史会议，拿到 meeting_ids
-lark-cli vc +search --query "周会" --start yesterday
+lark-cli vc +search --query "周会" --start 2026-03-10
 
 # 第 2 步：使用上一步返回的 meeting_ids 查询录制，拿到 minute_tokens
 lark-cli vc +recording --meeting-ids <ids>

skills/lark-vc/references/lark-vc-search.md

@@ -88,7 +88,17 @@ lark-cli vc +search --query "周会" --format json
 
 当返回 `has_more=true` 时，使用响应中的 `page_token` 配合 `--page-token` 获取下一页结果。
 
-### 5. 日期型 `--end` 包含当天整天
+### 5. 机器人可同时加入多个会议
+
+机器人支持同时加入多个正在进行中的会议；加入新会议前，不需要先退出已经在会中的其他会议。
+
+这意味着：
+
+- 不要假设 bot 一次只能在一个会议中
+- 如果用户要求 bot 再加入另一场会，可以直接继续执行对应的入会命令
+- 只有在用户明确要求结束某一场会中的 bot 参会时，才调用对应的离会命令
+
+### 6. 日期型 `--end` 包含当天整天
 
 当 `--end` 传入的是仅日期格式（如 `2026-03-10`）时，CLI 会将它解释为当天 `23:59:59`，而不是当天 `00:00:00`。
 

skills/lark-whiteboard/SKILL.md

@@ -13,7 +13,7 @@ metadata:
 
 > [!IMPORTANT]
 > - 运行 `lark-cli --version`，确认可用，无需询问用户。
-> - 运行 `npx -y @larksuite/whiteboard-cli@^0.2.10 -v`，确认可用，无需询问用户。
+> - 运行 `npx -y @larksuite/whiteboard-cli@^0.2.11 -v`，确认可用，无需询问用户。
 
 **CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理**
 
@@ -124,7 +124,7 @@ diagram.png           ← 渲染结果
 
 ```bash
 # 第一步：dry-run 探测
-npx -y @larksuite/whiteboard-cli@^0.2.10 -i <产物文件> --to openapi --format json \
+npx -y @larksuite/whiteboard-cli@^0.2.11 -i <产物文件> --to openapi --format json \
   | lark-cli whiteboard +update \
     --whiteboard-token <Token> \
     --source - --input_format raw \
@@ -132,7 +132,7 @@ npx -y @larksuite/whiteboard-cli@^0.2.10 -i <产物文件> --to openapi --format
     --overwrite --dry-run --as user
 
 # 第二步：确认后执行
-npx -y @larksuite/whiteboard-cli@^0.2.10 -i <产物文件> --to openapi --format json \
+npx -y @larksuite/whiteboard-cli@^0.2.11 -i <产物文件> --to openapi --format json \
   | lark-cli whiteboard +update \
     --whiteboard-token <Token> \
     --source - --input_format raw \

skills/lark-whiteboard/references/image.md

@@ -4,20 +4,24 @@
 
 ## 概述
 
-画板 DSL 支持 `type: 'image'` 节点，但图片不能直接使用 URL，必须先上传到飞书获取 **media token**，然后在 DSL 中引用。
+画板 DSL 支持 `type: 'image'` 节点，但图片不能直接使用 URL 或其他域的 token，**必须先上传到目标画板获取 `whiteboard` 域 media token**，然后在 DSL 中引用。
 
-**关键约束**：
-- 图片 token 必须通过 `docs +media-upload --parent-type whiteboard` 上传获取
-- 图片必须上传到**目标画板**（`--parent-node` 设为目标画板 token），跨画板的 token 不可用
-- `drive +upload` 获取的 Drive file token **不能**用于画板图片节点
+**核心规则**：不管图片从哪来（本地文件、URL、文档中的 `docx_image` token、其他域的 Drive token），都必须通过 `docs +media-upload --parent-type whiteboard --parent-node <目标画板token>` 上传，拿到画板专属的 media token 后才能在 DSL 中使用。直接使用非 `whiteboard` 域的 token 会导致画板 API 报 500（错误码 2891001）或图片在文档中消失。
 
 ## Step 0：图片准备流程
 
-### 1. 下载图片
+### 1. 获取图片到本地
 
-用 `curl` 下载图片到本地。**必须使用能根据关键词返回相关图片的图片源**。
+根据图片来源选择对应方式：
 
-**推荐图片源**：
+| 图片来源 | 获取方式 |
+|---------|---------|
+| 本地文件 | 直接使用 |
+| 网络 URL | `curl -L -o photo.jpg "<URL>"` |
+| 文档中的图片 token | `lark-cli docs +media-download --token <token> --output ./photo.png` |
+| 其他域的 Drive token | `lark-cli docs +media-download --token <token> --output ./photo.png` |
+
+**图片源选择（需要搜索图片时）**：
 
 | 图片源类型 | 说明 |
 |-------|------|
@@ -29,11 +33,6 @@
 - **关键词搜索**：支持按关键词搜索并返回相关图片，确保图片内容与主题匹配
 - **内容丰富**：图库图片种类多、数量大，能覆盖常见主题（宠物、美食、景点、产品等）
 
-```bash
-curl -L -o photo1.jpg "<图片URL>"
-curl -L -o photo2.jpg "<图片URL>"
-```
-
 **严禁使用随机占位图服务**：某些图库仅提供随机占位图，URL 中的关键词参数不会影响返回的图片内容，下载的图片与主题完全无关。
 
 ### 2. 校验图片
@@ -74,7 +73,8 @@ lark-cli docs +media-upload --file ./photo3.jpg --parent-type whiteboard --paren
 
 | 错误现象 | 原因 | 解决 |
 |---------|------|------|
-| 画板 API 返回 500（2891001） | 使用了 Drive file token 而非 media token | 改用 `docs +media-upload --parent-type whiteboard` |
+| 画板 API 返回 500（2891001） | 使用了非 `whiteboard` 域 token（如 `docx_image`、Drive file token） | 下载图片后用 `docs +media-upload --parent-type whiteboard` 重新上传 |
 | 画板 API 返回 500 | 图片上传到了其他画板 | 重新上传到目标画板 |
+| 画板在文档中图片消失 | 图片 token 的资源域与画板不匹配 | 确保图片通过 `--parent-type whiteboard --parent-node <画板token>` 上传 |
 | 图片裂开/无法显示 | token 无效或已过期 | 重新上传获取新 token |
 | 图片内容与主题无关 | 使用了随机占位图服务 | 改用免费版权图库服务 |

skills/lark-whiteboard/references/lark-whiteboard-update.md

@@ -74,7 +74,7 @@ whiteboard-cli 工具的具体用法请参考 [§ 渲染 & 写入画板](../SKIL
 
 ```bash
 # 使用 whiteboard-cli 生成 OpenAPI 格式并通过管道传递
-npx -y @larksuite/whiteboard-cli@^0.2.10 -i <产物文件> --to openapi --format json \
+npx -y @larksuite/whiteboard-cli@^0.2.11 -i <产物文件> --to openapi --format json \
   | lark-cli whiteboard +update \
     --whiteboard-token <画板Token> \
     --source - --input_format raw \
@@ -88,7 +88,7 @@ whiteboard-cli 工具的具体用法请参考 [§ 渲染 & 写入画板](../SKIL
 
 ```bash
 # 生成 OpenAPI 格式到文件
-npx -y @larksuite/whiteboard-cli@^0.2.10 -i <DSL 文件> --to openapi --format json -o ./temp.json
+npx -y @larksuite/whiteboard-cli@^0.2.11 -i <DSL 文件> --to openapi --format json -o ./temp.json
 
 # 从文件读取并更新
 lark-cli whiteboard +update \