fix(slides): lint 放行并识别 embed 嵌入元素

schema 已支持 <embed> 嵌入 SVG 协议,但版式准出 lint 未同步: - SXSD 标签校验仅在 <whiteboard> 祖先下放行 SVG 命名空间子树, <embed> 内的合法 <svg> 会被判 sxsd_unsupported_tag 并 block - extract_elements 不提取 embed,仅含 embed 的页面被误报 blank_slide,也不参与重叠/密度检查修复:SVG 子树白名单加入 embed 祖先并更新 svg 报错 hint;embed 进入元素提取以及 slide_content_visual_bbox、is_large_visual_child 的视觉元素集合。新增 embed 内 svg 放行、embed-only 不报空白页两个测试,extract_elements 测试覆盖 embed。 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
feat(slides): 支持嵌入 SVG 内容协议
2026-08-03 08:32:46 +08:00 · 2026-07-29 11:59:13 +08:00 · 2026-07-29 11:30:29 +08:00 · 2026-07-29 10:55:05 +08:00 · 2026-07-28 21:45:00 +08:00 · 2026-07-28 21:02:37 +08:00
34 changed files with 1929 additions and 326 deletions
--- a/.github/workflows/semantic-review.yml
+++ b/.github/workflows/semantic-review.yml
@@ -25,19 +25,16 @@ jobs:
        with:
          script: |
            const run = context.payload.workflow_run;
-            if (run.name !== "CI") throw new Error(`unexpected workflow name: ${run.name}`);
-            let workflowPath = run.path || "";
-            if (!workflowPath) {
-              const workflowId = Number(run.workflow_id || 0);
-              if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
-              const { data: workflow } = await github.rest.actions.getWorkflow({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                workflow_id: workflowId,
-              });
-              workflowPath = workflow.path || "";
-            }
-            if (workflowPath !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflowPath}`);
+            const workflowId = Number(run.workflow_id || 0);
+            if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
+            const { data: workflow } = await github.rest.actions.getWorkflow({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              workflow_id: workflowId,
+            });
+            if (workflow.name !== "CI") throw new Error(`unexpected workflow name: ${workflow.name}`);
+            if (workflow.path !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflow.path}`);
+            if (run.path && run.path !== workflow.path) throw new Error(`workflow path mismatch: ${run.path}`);
            if (run.event !== "pull_request") throw new Error(`unexpected event: ${run.event}`);
            if (run.repository.id !== context.payload.repository.id) throw new Error("repository id mismatch");
            if (run.repository.full_name !== context.payload.repository.full_name) throw new Error("repository name mismatch");
@@ -253,19 +250,16 @@ jobs:
        with:
          script: |
            const run = context.payload.workflow_run;
-            if (run.name !== "CI") throw new Error(`unexpected workflow name: ${run.name}`);
-            let workflowPath = run.path || "";
-            if (!workflowPath) {
-              const workflowId = Number(run.workflow_id || 0);
-              if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
-              const { data: workflow } = await github.rest.actions.getWorkflow({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                workflow_id: workflowId,
-              });
-              workflowPath = workflow.path || "";
-            }
-            if (workflowPath !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflowPath}`);
+            const workflowId = Number(run.workflow_id || 0);
+            if (!Number.isInteger(workflowId) || workflowId <= 0) throw new Error("missing workflow id");
+            const { data: workflow } = await github.rest.actions.getWorkflow({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              workflow_id: workflowId,
+            });
+            if (workflow.name !== "CI") throw new Error(`unexpected workflow name: ${workflow.name}`);
+            if (workflow.path !== ".github/workflows/ci.yml") throw new Error(`unexpected workflow path: ${workflow.path}`);
+            if (run.path && run.path !== workflow.path) throw new Error(`workflow path mismatch: ${run.path}`);
            if (run.event !== "pull_request") throw new Error(`unexpected event: ${run.event}`);
            if (run.conclusion !== "success") throw new Error(`unexpected conclusion: ${run.conclusion}`);
            if (run.repository.id !== context.payload.repository.id) throw new Error("repository id mismatch");
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,40 @@

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

+## [v1.0.79] - 2026-07-28
+
+### Features
+
+- **slides**: update xsd (#2067)
+
+### Bug Fixes
+
+- **ci**: validate static workflow identity (#2015)
+- **sheets**: recognize OFL0X local office tokens (#2063)
+
+### Documentation
+
+- **calendar**: clarify identity selection by event ownership (#2071)
+- **slides**: add formula inline element syntax to quick-ref (#2077)
+
+## [v1.0.78] - 2026-07-27
+
+### Features
+
+- event description support rich text (#1975)
+
+### Bug Fixes
+
+- **slides**: restrict canvas overflow checks
+- **slides**: upgrade text overflow to error above 10px threshold
+- **slides**: detect letterSpacing-driven text overflow
+- **slides**: downgrade background-decoration text overflow to info
+- **slides**: allow chartParsedValues roundtrip tag
+- refine character width estimation for lark-slides text lint
+- **slides**: preserve info lint severity
+- **slides**: text may over flow shape
+- exempt ghost text from slides lint
+
 ## [v1.0.77] - 2026-07-24

 ### Features
@@ -1667,6 +1701,8 @@ Bundled AI agent skills for intelligent assistance:
 - Bilingual documentation (English & Chinese).
 - CI/CD pipelines: linting, testing, coverage reporting, and automated releases.

+[v1.0.79]: https://github.com/larksuite/cli/releases/tag/v1.0.79
+[v1.0.78]: https://github.com/larksuite/cli/releases/tag/v1.0.78
 [v1.0.77]: https://github.com/larksuite/cli/releases/tag/v1.0.77
 [v1.0.75]: https://github.com/larksuite/cli/releases/tag/v1.0.75
 [v1.0.74]: https://github.com/larksuite/cli/releases/tag/v1.0.74
--- a/package-lock.json
+++ b/package-lock.json
@@ -1,12 +1,12 @@
 {
  "name": "@larksuite/cli",
-  "version": "1.0.77",
+  "version": "1.0.79",
  "lockfileVersion": 3,
  "requires": true,
  "packages": {
    "": {
      "name": "@larksuite/cli",
-      "version": "1.0.77",
+      "version": "1.0.79",
      "cpu": [
        "x64",
        "arm64",
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
  "name": "@larksuite/cli",
-  "version": "1.0.77",
+  "version": "1.0.79",
  "description": "The official CLI for Lark/Feishu open platform",
  "bin": {
    "lark-cli": "scripts/run.js"
--- a/scripts/semantic-review-workflow.test.sh
+++ b/scripts/semantic-review-workflow.test.sh
@@ -176,7 +176,15 @@ if ! grep -Fq "if: always() && github.event.workflow_run.conclusion == 'success'
  exit 1
 fi

-require_in_step "$summary_verify_step" 'workflowPath !== ".github/workflows/ci.yml"' "PR quality summary must verify the triggering workflow path"
+if grep -Fq 'run.name !== "CI"' "$workflow"; then
+  echo "semantic-review must not use the dynamic workflow run name as workflow identity" >&2
+  exit 1
+fi
+
+require_in_step "$summary_verify_step" 'github.rest.actions.getWorkflow' "PR quality summary must resolve static workflow metadata"
+require_in_step "$summary_verify_step" 'workflow.name !== "CI"' "PR quality summary must verify the static workflow name"
+require_in_step "$summary_verify_step" 'workflow.path !== ".github/workflows/ci.yml"' "PR quality summary must verify the static workflow path"
+require_in_step "$summary_verify_step" 'run.path && run.path !== workflow.path' "PR quality summary must reject workflow path metadata mismatches"
 require_in_step "$summary_verify_step" 'run.event !== "pull_request"' "PR quality summary must only handle pull_request workflow_run events"
 require_in_step "$summary_verify_step" 'run.repository.id !== context.payload.repository.id' "PR quality summary must verify workflow_run repository id"
 require_in_step "$summary_verify_step" 'const targetHeadSha = run.head_sha' "PR quality summary must use the CI run head SHA as the verified PR head"
@@ -201,7 +209,10 @@ require_in_step "$summary_publish_step" 'CI_QUALITY_SUMMARY_BASE_SHA' "PR qualit
 require_in_step "$summary_publish_step" 'CI_QUALITY_SUMMARY_RUN_ID' "PR quality summary publisher must receive verified workflow run id"
 require_in_step "$summary_publish_step" 'require("./scripts/ci-quality-summary-publish.js")' "PR quality summary publisher must use the shared CI publisher script"

-require_in_step "$verify_step" 'workflowPath !== ".github/workflows/ci.yml"' "semantic-review must verify the triggering workflow path"
+require_in_step "$verify_step" 'github.rest.actions.getWorkflow' "semantic-review must resolve static workflow metadata"
+require_in_step "$verify_step" 'workflow.name !== "CI"' "semantic-review must verify the static workflow name"
+require_in_step "$verify_step" 'workflow.path !== ".github/workflows/ci.yml"' "semantic-review must verify the static workflow path"
+require_in_step "$verify_step" 'run.path && run.path !== workflow.path' "semantic-review must reject workflow path metadata mismatches"
 require_in_step "$verify_step" 'run.repository.id !== context.payload.repository.id' "semantic-review must verify workflow_run repository id"
 require_in_step "$verify_step" 'run.event !== "pull_request"' "semantic-review must only handle pull_request workflow_run events"
 require_in_step "$verify_step" 'run.conclusion !== "success"' "semantic-review must only consume successful CI runs"
--- a/shortcuts/calendar/calendar_agenda.go
+++ b/shortcuts/calendar/calendar_agenda.go
@@ -250,7 +250,7 @@ var CalendarAgenda = common.Shortcut{
 					}
 				}

-				backfillDescriptionRich(e)
+				collapseDescription(e)

 				filtered = append(filtered, e)
 			}
--- a/shortcuts/calendar/calendar_create.go
+++ b/shortcuts/calendar/calendar_create.go
@@ -32,8 +32,8 @@ func buildEventData(runtime *common.RuntimeContext, startTs, endTs string) map[s
 	if rrule := runtime.Str("rrule"); rrule != "" {
 		eventData["recurrence"] = rrule
 	}
-	if descriptionRich := descriptionRichToSend(runtime); descriptionRich != "" {
-		eventData["description_rich"] = descriptionRich
+	if description := descriptionToSend(runtime); description != "" {
+		eventData["description_rich"] = description
 	}
 	return eventData
 }
@@ -120,8 +120,7 @@ var CalendarCreate = common.Shortcut{
 		{Name: "summary", Desc: "event title"},
 		{Name: "start", Desc: "start time (ISO 8601)", Required: true},
 		{Name: "end", Desc: "end time (ISO 8601)", Required: true},
-		{Name: "description", Desc: "deprecated: plain-text description; use --description-rich (Markdown) instead", Hidden: true},
-		{Name: "description-rich", Desc: "event description as Markdown (@file or - for stdin); the unified description field. Supports bold/italic/underline/strikethrough, links, headings (`#`..`###`), blockquotes (`>`), ordered/unordered lists, horizontal rules (`---`), GFM tables, and images (`![name](url)`; a remote URL is used as-is, and a local image path relative to and inside the current working directory is auto-uploaded to Lark drive and rendered inline — absolute/out-of-cwd paths are rejected). A Lark doc URL (bare or as a Markdown link) is auto-resolved to an inline doc-mention chip showing its title. Inside a GFM table cell, stack multiple lines with `<br>`; each line may itself be an ordered/unordered list item, image or styled text (e.g. `1. a<br>2. b`, `- x<br>- y`, `![p](url)<br>**bold**`).", Input: []string{common.File, common.Stdin}},
+		{Name: "description", Desc: "event description as Markdown (@file or - for stdin); the unified description field. Supports bold/italic/underline/strikethrough, links, headings (`#`..`###`), blockquotes (`>`), ordered/unordered lists, horizontal rules (`---`), GFM tables, and images (`![name](url)`; a remote URL is used as-is, and a local image path relative to and inside the current working directory is auto-uploaded to Lark drive and rendered inline — absolute/out-of-cwd paths are rejected). A Lark doc URL (bare or as a Markdown link) is auto-resolved to an inline doc-mention chip showing its title. Inside a GFM table cell, stack multiple lines with `<br>`; each line may itself be an ordered/unordered list item, image or styled text (e.g. `1. a<br>2. b`, `- x<br>- y`, `![p](url)<br>**bold**`).", Input: []string{common.File, common.Stdin}},
 		{Name: "attendee-ids", Desc: "attendee IDs, comma-separated (supports user ou_, chat oc_, room omm_)"},
 		{Name: "calendar-id", Desc: "calendar ID (default: primary)"},
 		{Name: "rrule", Desc: "recurrence rule (rfc5545)"},
@@ -234,7 +233,7 @@ var CalendarCreate = common.Shortcut{
 		if err != nil {
 			return errs.NewValidationError(errs.SubtypeInvalidArgument, "--end: %v", err).WithParam("--end")
 		}
-		if err := resolveDescriptionRichImages(runtime, calendarId); err != nil {
+		if err := resolveDescriptionImages(runtime, calendarId); err != nil {
 			return err
 		}

--- a/shortcuts/calendar/calendar_get.go
+++ b/shortcuts/calendar/calendar_get.go
@@ -170,7 +170,7 @@ func buildCalendarEventOutput(event *calendarEvent) (map[string]interface{}, err
 	if status, _ := out["status"].(string); status != "cancelled" {
 		delete(out, "status")
 	}
-	backfillDescriptionRich(out)
+	collapseDescription(out)
 	return out, nil
 }

--- a/shortcuts/calendar/calendar_test.go
+++ b/shortcuts/calendar/calendar_test.go
@@ -988,8 +988,9 @@ func TestUpdate_PatchEventOnly(t *testing.T) {
 	if err := json.Unmarshal(stub.CapturedBody, &body); err != nil {
 		t.Fatalf("unmarshal captured patch body: %v", err)
 	}
-	// The deprecated, hidden --description folds into description_rich; the CLI
-	// never sends the plain description field (mutually exclusive downstream).
+	// --description is the unified field, treated as rich text and sent as
+	// description_rich; the CLI never sends the plain description field
+	// (mutually exclusive downstream).
 	if body["summary"] != "Updated Meeting" || body["description_rich"] != "Updated description" {
 		t.Fatalf("unexpected patch body: %#v", body)
 	}
@@ -1411,18 +1412,17 @@ func TestAgenda_UnifiesDescriptionRich(t *testing.T) {
 		t.Fatalf("unexpected error: %v", err)
 	}
 	out := stdout.String()
-	// Read keeps both fields: description (plain) and description_rich (rich).
-	if !strings.Contains(out, "\"description\": \"[测试]\\n友情提醒\"") {
-		t.Errorf("expected plain description retained, got: %s", out)
-	}
-	if !strings.Contains(out, "\"description_rich\": \"友情提醒\"") {
-		t.Errorf("expected rich description surfaced, got: %s", out)
+	// Read exposes a single unified description field: it carries the rich
+	// (Markdown) value when present, and the plain text otherwise. The internal
+	// description_rich key is never surfaced.
+	if !strings.Contains(out, "\"description\": \"友情提醒\"") {
+		t.Errorf("expected rich value surfaced under description, got: %s", out)
 	}
 	if !strings.Contains(out, "\"description\": \"just text\"") {
-		t.Errorf("expected plain description retained for plain-only event, got: %s", out)
+		t.Errorf("expected plain description surfaced for plain-only event, got: %s", out)
 	}
-	if !strings.Contains(out, "\"description_rich\": \"just text\"") {
-		t.Errorf("expected description_rich backfilled from plain, got: %s", out)
+	if strings.Contains(out, "description_rich") {
+		t.Errorf("description_rich must not appear in output, got: %s", out)
 	}
 }

@@ -3438,7 +3438,8 @@ func TestGet_Success_FlattensAndConvertsTimes(t *testing.T) {
 }

 func TestGet_UnifiesDescriptionRich(t *testing.T) {
-	// Read keeps both description (plain) and description_rich (rich).
+	// Read exposes a single unified description field carrying the rich value
+	// when present, and the plain text otherwise; description_rich is dropped.
 	t.Run("rich present", func(t *testing.T) {
 		f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
 		reg.Register(&httpmock.Stub{
@@ -3462,17 +3463,16 @@ func TestGet_UnifiesDescriptionRich(t *testing.T) {
 			t.Fatalf("unexpected error: %v", err)
 		}
 		out := stdout.String()
-		if !strings.Contains(out, "\"description\": \"[表格]\"") {
-			t.Errorf("expected plain description retained, got: %s", out)
+		if !strings.Contains(out, "| a | b |") {
+			t.Errorf("expected rich value surfaced under description, got: %s", out)
 		}
-		if !strings.Contains(out, "\"description_rich\":") {
-			t.Errorf("expected description_rich in output, got: %s", out)
+		if strings.Contains(out, "description_rich") {
+			t.Errorf("description_rich must not appear in output, got: %s", out)
 		}
 	})

-	// When only a plain description exists, description_rich is backfilled from it
-	// and the plain description is still returned.
-	t.Run("only plain backfills rich", func(t *testing.T) {
+	// When only a plain description exists, it is surfaced under description.
+	t.Run("only plain surfaces under description", func(t *testing.T) {
 		f, stdout, _, reg := cmdutil.TestFactory(t, defaultConfig())
 		reg.Register(&httpmock.Stub{
 			Method: "GET",
@@ -3495,10 +3495,10 @@ func TestGet_UnifiesDescriptionRich(t *testing.T) {
 		}
 		out := stdout.String()
 		if !strings.Contains(out, "\"description\": \"just text\"") {
-			t.Errorf("expected plain description retained, got: %s", out)
+			t.Errorf("expected plain description surfaced, got: %s", out)
 		}
-		if !strings.Contains(out, "\"description_rich\": \"just text\"") {
-			t.Errorf("expected description_rich backfilled from plain, got: %s", out)
+		if strings.Contains(out, "description_rich") {
+			t.Errorf("description_rich must not appear in output, got: %s", out)
 		}
 	})
 }
--- a/shortcuts/calendar/calendar_update.go
+++ b/shortcuts/calendar/calendar_update.go
@@ -29,8 +29,7 @@ var CalendarUpdate = common.Shortcut{
 		{Name: "event-id", Desc: "event ID to update", Required: true},
 		{Name: "calendar-id", Desc: "calendar ID (default: primary)"},
 		{Name: "summary", Desc: "event title"},
-		{Name: "description", Desc: "deprecated: plain-text description; use --description-rich (Markdown) instead", Hidden: true},
-		{Name: "description-rich", Desc: "event description as Markdown (@file or - for stdin); the unified description field. Supports bold/italic/underline/strikethrough, links, headings (`#`..`###`), blockquotes (`>`), ordered/unordered lists, horizontal rules (`---`), GFM tables, and images (`![name](url)`; a remote URL is used as-is, and a local image path relative to and inside the current working directory is auto-uploaded to Lark drive and rendered inline — absolute/out-of-cwd paths are rejected). A Lark doc URL (bare or as a Markdown link) is auto-resolved to an inline doc-mention chip showing its title. Inside a GFM table cell, stack multiple lines with `<br>`; each line may itself be an ordered/unordered list item, image or styled text (e.g. `1. a<br>2. b`, `- x<br>- y`, `![p](url)<br>**bold**`).", Input: []string{common.File, common.Stdin}},
+		{Name: "description", Desc: "event description as Markdown (@file or - for stdin); the unified description field. Supports bold/italic/underline/strikethrough, links, headings (`#`..`###`), blockquotes (`>`), ordered/unordered lists, horizontal rules (`---`), GFM tables, and images (`![name](url)`; a remote URL is used as-is, and a local image path relative to and inside the current working directory is auto-uploaded to Lark drive and rendered inline — absolute/out-of-cwd paths are rejected). A Lark doc URL (bare or as a Markdown link) is auto-resolved to an inline doc-mention chip showing its title. Inside a GFM table cell, stack multiple lines with `<br>`; each line may itself be an ordered/unordered list item, image or styled text (e.g. `1. a<br>2. b`, `- x<br>- y`, `![p](url)<br>**bold**`). Passing an empty string clears the description.", Input: []string{common.File, common.Stdin}},
 		{Name: "start", Desc: "new start time (ISO 8601); requires --end"},
 		{Name: "end", Desc: "new end time (ISO 8601); requires --start"},
 		{Name: "rrule", Desc: "recurrence rule (rfc5545)"},
@@ -72,7 +71,7 @@ func validateCalendarUpdate(runtime *common.RuntimeContext) error {
 		return err
 	}
 	if !hasCalendarUpdateOperation(runtime) {
-		return errs.NewValidationError(errs.SubtypeInvalidArgument, "nothing to update: specify at least one of --summary, --description, --description-rich, --start/--end, --rrule, --add-attendee-ids, or --remove-attendee-ids")
+		return errs.NewValidationError(errs.SubtypeInvalidArgument, "nothing to update: specify at least one of --summary, --description, --start/--end, --rrule, --add-attendee-ids, or --remove-attendee-ids")
 	}
 	return nil
 }
@@ -114,10 +113,7 @@ func buildCalendarUpdateEventData(runtime *common.RuntimeContext) (map[string]in
 		body["summary"] = runtime.Str("summary")
 		hasFields = true
 	}
-	if runtime.Cmd.Flags().Changed("description-rich") {
-		body["description_rich"] = runtime.Str("description-rich")
-		hasFields = true
-	} else if runtime.Cmd.Flags().Changed("description") {
+	if runtime.Cmd.Flags().Changed("description") {
 		body["description_rich"] = runtime.Str("description")
 		hasFields = true
 	}
@@ -362,11 +358,8 @@ func executeCalendarUpdate(ctx context.Context, runtime *common.RuntimeContext)
 		return errs.NewValidationError(errs.SubtypeInvalidArgument, "specify --event-id").WithParam("--event-id")
 	}

-	// Upload any local images referenced in --description-rich and rewrite them
-	// to drive URLs before the description is sent (the service cannot read
-	// local files).
-	if runtime.Cmd.Flags().Changed("description-rich") {
-		if err := resolveDescriptionRichImages(runtime, calendarID); err != nil {
+	if runtime.Cmd.Flags().Changed("description") {
+		if err := resolveDescriptionImages(runtime, calendarID); err != nil {
 			return err
 		}
 	}
@@ -443,19 +436,10 @@ func calendarUpdateResult(eventID string, event map[string]interface{}, addedCou
 	if summary, _ := event["summary"].(string); summary != "" {
 		result["summary"] = summary
 	}
-	// Surface both description fields on read: description holds plain text,
-	// description_rich holds the rich (Markdown) version, backfilled from plain
-	// when the service returned no rich value.
-	description, _ := event["description"].(string)
-	if description != "" {
-		result["description"] = description
-	}
-	descriptionRich, _ := event["description_rich"].(string)
-	if descriptionRich == "" {
-		descriptionRich = description
-	}
-	if descriptionRich != "" {
-		result["description_rich"] = descriptionRich
+	if rich, _ := event["description_rich"].(string); rich != "" {
+		result["description"] = rich
+	} else if plain, _ := event["description"].(string); plain != "" {
+		result["description"] = plain
 	}
 	if start := formatCalendarEventTime(event["start_time"]); start != "" {
 		result["start"] = start
--- a/shortcuts/calendar/description_rich_images.go
+++ b/shortcuts/calendar/description_rich_images.go
@@ -27,8 +27,8 @@ const calendarMediaParentType = "calendar"

 var markdownImageRe = regexp.MustCompile(`!\[([^\]]*)\]\(([^)]*)\)`)

-func resolveDescriptionRichImages(runtime *common.RuntimeContext, calendarID string) error {
-	md := runtime.Str("description-rich")
+func resolveDescriptionImages(runtime *common.RuntimeContext, calendarID string) error {
+	md := runtime.Str("description")
 	if md == "" || !strings.Contains(md, "![") {
 		return nil
 	}
@@ -37,8 +37,8 @@ func resolveDescriptionRichImages(runtime *common.RuntimeContext, calendarID str
 		return err
 	}
 	if changed {
-		if err := runtime.Cmd.Flags().Set("description-rich", rewritten); err != nil {
-			return errs.NewInternalError(errs.SubtypeUnknown, "failed to update --description-rich after image upload: %v", err).WithCause(err)
+		if err := runtime.Cmd.Flags().Set("description", rewritten); err != nil {
+			return errs.NewInternalError(errs.SubtypeUnknown, "failed to update --description after image upload: %v", err).WithCause(err)
 		}
 	}
 	return nil
@@ -85,8 +85,8 @@ func resolveLocalImage(runtime *common.RuntimeContext, calendarID, src, alt stri
 	safePath, err := validate.SafeInputPath(localPath)
 	if err != nil {
 		return "", errs.NewValidationError(errs.SubtypeInvalidArgument,
-			"--description-rich image %q could not be read: %v", src, err).
-			WithParam("--description-rich").
+			"--description image %q could not be read: %v", src, err).
+			WithParam("--description").
 			WithHint("reference local images by a path inside the current working directory (e.g. ./images/pic.png; cd there first), or use an already-uploaded Lark image URL").
 			WithCause(err)
 	}
@@ -157,7 +157,7 @@ func localImagePath(src string) string {
 }

 func buildCalendarImagePreviewURL(brand core.LarkBrand, fileToken string, width, height int, size int64) string {
-	host := "internal-api-drive-stream.larkoffice.com"
+	host := "internal-api-drive-stream.feishu.cn"
 	if brand == core.BrandLark {
 		host = "internal-api-drive-stream.larksuite.com"
 	}
--- a/shortcuts/calendar/description_rich_images_test.go
+++ b/shortcuts/calendar/description_rich_images_test.go
@@ -68,7 +68,7 @@ func TestBuildCalendarImagePreviewURL(t *testing.T) {
 		brand    core.LarkBrand
 		hostFrag string
 	}{
-		{core.BrandFeishu, "larkoffice"},
+		{core.BrandFeishu, "feishu.cn"},
 		{core.BrandLark, "larksuite"},
 	} {
 		raw := buildCalendarImagePreviewURL(tc.brand, "boxcnTOKEN123", 416, 306, 142568)
@@ -158,7 +158,7 @@ func TestCreate_UploadsLocalDescriptionImage(t *testing.T) {
 		"--start", "2025-03-21T00:00:00+08:00",
 		"--end", "2025-03-21T01:00:00+08:00",
 		"--calendar-id", "cal_test123",
-		"--description-rich", "![pic](./pic.png)",
+		"--description", "![pic](./pic.png)",
 		"--as", "bot",
 	}, f, stdout)
 	if runErr != nil {
@@ -233,7 +233,7 @@ func TestCreate_LocalImageCarriesDimensions(t *testing.T) {
 		"--start", "2025-03-21T00:00:00+08:00",
 		"--end", "2025-03-21T01:00:00+08:00",
 		"--calendar-id", "cal_test123",
-		"--description-rich", "![pic](./pic.png)",
+		"--description", "![pic](./pic.png)",
 		"--as", "bot",
 	}, f, stdout)
 	if runErr != nil {
@@ -253,7 +253,7 @@ func TestCreate_LocalImageCarriesDimensions(t *testing.T) {
 }

 // TestCreate_LocalImageAbsolutePathRejected verifies an out-of-cwd absolute path
-// yields a typed --description-rich validation error before any API call.
+// yields a typed --description validation error before any API call.
 func TestCreate_LocalImageAbsolutePathRejected(t *testing.T) {
 	f, stdout, _, _ := cmdutil.TestFactory(t, defaultConfig())

@@ -263,7 +263,7 @@ func TestCreate_LocalImageAbsolutePathRejected(t *testing.T) {
 		"--start", "2025-03-21T00:00:00+08:00",
 		"--end", "2025-03-21T01:00:00+08:00",
 		"--calendar-id", "cal_test123",
-		"--description-rich", "![p](/etc/hosts)",
+		"--description", "![p](/etc/hosts)",
 		"--as", "bot",
 	}, f, stdout)
 	if runErr == nil {
@@ -273,7 +273,7 @@ func TestCreate_LocalImageAbsolutePathRejected(t *testing.T) {
 	if !errors.As(runErr, &ve) {
 		t.Fatalf("expected *errs.ValidationError, got %T: %v", runErr, runErr)
 	}
-	if ve.Param != "--description-rich" {
-		t.Errorf("param = %q, want --description-rich", ve.Param)
+	if ve.Param != "--description" {
+		t.Errorf("param = %q, want --description", ve.Param)
 	}
 }
--- a/shortcuts/calendar/helpers.go
+++ b/shortcuts/calendar/helpers.go
@@ -30,20 +30,23 @@ func resolveStartEnd(runtime *common.RuntimeContext) (string, string) {
 	return startInput, endInput
 }

-func backfillDescriptionRich(event map[string]interface{}) {
+func collapseDescription(event map[string]interface{}) {
 	if event == nil {
 		return
 	}
-	if descRich, _ := event["description_rich"].(string); descRich == "" {
-		if desc, _ := event["description"].(string); desc != "" {
-			event["description_rich"] = desc
-		}
+	rich, _ := event["description_rich"].(string)
+	plain, _ := event["description"].(string)
+	delete(event, "description_rich")
+	switch {
+	case rich != "":
+		event["description"] = rich
+	case plain != "":
+		event["description"] = plain
+	default:
+		delete(event, "description")
 	}
 }
-func descriptionRichToSend(runtime *common.RuntimeContext) string {
-	if v := runtime.Str("description-rich"); v != "" {
-		return v
-	}
+func descriptionToSend(runtime *common.RuntimeContext) string {
 	return runtime.Str("description")
 }

--- a/shortcuts/sheets/backward/lark_sheets_float_images.go
+++ b/shortcuts/sheets/backward/lark_sheets_float_images.go
@@ -17,9 +17,10 @@ import (
 )

 // Drive media parent_type values for uploading an image into a spreadsheet.
-// Native spreadsheets use "sheet_image"; imported "office" spreadsheets carry a
-// synthetic token prefixed with "fake_office_" (being renamed to
-// "local_office_") and the backend requires "office_sheet_file" instead.
+// Native spreadsheets use "sheet_image"; imported "office" spreadsheets use a
+// legacy synthetic-token prefix or a 28-character token whose interleaved
+// product/region marker is "OFL0X". The backend requires
+// "office_sheet_file" for those imported spreadsheets.
 const (
 	sheetImageParentType      = "sheet_image"
 	officeSheetFileParentType = "office_sheet_file"
@@ -27,22 +28,37 @@ const (
 	localOfficePrefix         = "local_office_"
 )

-// officePrefixes are the synthetic token prefixes an imported "office"
-// spreadsheet may carry. The prefix is being renamed from "fake_office_" to
-// "local_office_"; accept either so image uploads keep working across the
-// rename.
+// officePrefixes are the legacy synthetic token prefixes an imported "office"
+// spreadsheet may carry.
 var officePrefixes = []string{fakeOfficePrefix, localOfficePrefix}

-// sheetMediaParentType returns the drive media parent_type to use when
-// uploading an image whose parent_node is spreadsheetToken, mapping either the
-// "fake_office_" or "local_office_" imported-spreadsheet token prefix to
-// "office_sheet_file".
-func sheetMediaParentType(spreadsheetToken string) string {
+func isOfficeSpreadsheet(spreadsheetToken string) bool {
 	for _, prefix := range officePrefixes {
 		if strings.HasPrefix(spreadsheetToken, prefix) {
-			return officeSheetFileParentType
+			return true
 		}
 	}
+	if len(spreadsheetToken) != 28 {
+		return false
+	}
+	// The five-character marker occupies positions 5, 10, 15, 20, and 25
+	// (1-based) in the interleaved token.
+	marker := []byte{
+		spreadsheetToken[4],
+		spreadsheetToken[9],
+		spreadsheetToken[14],
+		spreadsheetToken[19],
+		spreadsheetToken[24],
+	}
+	return string(marker) == "OFL0X"
+}
+
+// sheetMediaParentType returns the drive media parent_type to use when
+// uploading an image whose parent_node is spreadsheetToken.
+func sheetMediaParentType(spreadsheetToken string) string {
+	if isOfficeSpreadsheet(spreadsheetToken) {
+		return officeSheetFileParentType
+	}
 	return sheetImageParentType
 }

--- a/shortcuts/sheets/backward/lark_sheets_sheet_media_upload_test.go
+++ b/shortcuts/sheets/backward/lark_sheets_sheet_media_upload_test.go
@@ -105,7 +105,7 @@ func TestSheetMediaUploadDryRunSmallFileOfficeParentType(t *testing.T) {
 	f, stdout, _, _ := cmdutil.TestFactory(t, sheetsTestConfig())
 	err := mountAndRunSheets(t, SheetMediaUpload, []string{
 		"+media-upload",
-		"--spreadsheet-token", "fake_office_abc123",
+		"--spreadsheet-token", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
 		"--file", "img.png",
 		"--dry-run", "--as", "user",
 	}, f, stdout)
@@ -117,10 +117,10 @@ func TestSheetMediaUploadDryRunSmallFileOfficeParentType(t *testing.T) {
 		t.Fatalf("dry-run should use upload_all for small file, got: %s", out)
 	}
 	if !strings.Contains(out, `"office_sheet_file"`) {
-		t.Fatalf("dry-run should include parent_type=office_sheet_file for fake_office_ token, got: %s", out)
+		t.Fatalf("dry-run should include parent_type=office_sheet_file for interleaved OFL0X token, got: %s", out)
 	}
 	if strings.Contains(out, `"sheet_image"`) {
-		t.Fatalf("dry-run must not emit sheet_image for fake_office_ token, got: %s", out)
+		t.Fatalf("dry-run must not emit sheet_image for interleaved OFL0X token, got: %s", out)
 	}
 }

@@ -239,7 +239,7 @@ func TestSheetMediaUploadExecuteSuccess(t *testing.T) {
 }

 // TestSheetMediaUploadExecuteOfficeParentType confirms that an imported
-// "office" spreadsheet (token prefixed with "fake_office_") uploads with
+// "office" spreadsheet (token carrying the interleaved "OFL0X" marker) uploads with
 // parent_type=office_sheet_file instead of the native sheet_image.
 func TestSheetMediaUploadExecuteOfficeParentType(t *testing.T) {
 	dir := t.TempDir()
@@ -259,7 +259,7 @@ func TestSheetMediaUploadExecuteOfficeParentType(t *testing.T) {
 	}
 	reg.Register(stub)

-	const officeToken = "fake_office_abc123"
+	const officeToken = "aaaaOaaaaFaaaaLaaaa0aaaaXaaa"
 	err := mountAndRunSheets(t, SheetMediaUpload, []string{
 		"+media-upload",
 		"--spreadsheet-token", officeToken,
--- a/shortcuts/sheets/helpers.go
+++ b/shortcuts/sheets/helpers.go
@@ -53,9 +53,10 @@ func sheetsInputStatError(flag string, err error) error {
 }

 // Drive media parent_type values for uploading an image into a spreadsheet.
-// Native spreadsheets use "sheet_image"; imported "office" spreadsheets carry a
-// synthetic token prefixed with "fake_office_" (being renamed to
-// "local_office_") and the backend requires "office_sheet_file" instead.
+// Native spreadsheets use "sheet_image"; imported "office" spreadsheets use a
+// legacy synthetic-token prefix or a 28-character token whose interleaved
+// product/region marker is "OFL0X". The backend requires
+// "office_sheet_file" for those imported spreadsheets.
 const (
 	sheetImageParentType      = "sheet_image"
 	officeSheetFileParentType = "office_sheet_file"
@@ -63,21 +64,38 @@ const (
 	localOfficePrefix         = "local_office_"
 )

-// officePrefixes are the synthetic token prefixes an imported "office"
-// spreadsheet may carry. The prefix is being renamed from "fake_office_" to
-// "local_office_"; accept either so image uploads keep working across the
-// rename.
+// officePrefixes are the legacy synthetic token prefixes an imported "office"
+// spreadsheet may carry.
 var officePrefixes = []string{fakeOfficePrefix, localOfficePrefix}

+func isOfficeSpreadsheet(spreadsheetToken string) bool {
+	for _, prefix := range officePrefixes {
+		if strings.HasPrefix(spreadsheetToken, prefix) {
+			return true
+		}
+	}
+	if len(spreadsheetToken) != 28 {
+		return false
+	}
+	// The five-character marker occupies positions 5, 10, 15, 20, and 25
+	// (1-based) in the interleaved token.
+	marker := []byte{
+		spreadsheetToken[4],
+		spreadsheetToken[9],
+		spreadsheetToken[14],
+		spreadsheetToken[19],
+		spreadsheetToken[24],
+	}
+	return string(marker) == "OFL0X"
+}
+
 // sheetMediaParentType returns the drive media parent_type to use when
 // uploading an image whose parent_node is spreadsheetToken. It is the single
 // place that maps a spreadsheet token to its parent_type so every image-upload
 // entry point (and its dry-run preview) stays consistent.
 func sheetMediaParentType(spreadsheetToken string) string {
-	for _, prefix := range officePrefixes {
-		if strings.HasPrefix(spreadsheetToken, prefix) {
-			return officeSheetFileParentType
-		}
+	if isOfficeSpreadsheet(spreadsheetToken) {
+		return officeSheetFileParentType
 	}
 	return sheetImageParentType
 }
--- a/shortcuts/sheets/sheet_media_parent_type_test.go
+++ b/shortcuts/sheets/sheet_media_parent_type_test.go
@@ -25,8 +25,9 @@ import (

 // TestSheetMediaParentType pins the token→parent_type mapping that every
 // sheets image-upload entry point funnels through. Native spreadsheet tokens
-// use "sheet_image"; imported "office" spreadsheets carry a "fake_office_" or
-// "local_office_" synthetic token and must upload with "office_sheet_file".
+// use "sheet_image"; imported "office" spreadsheets use either a legacy
+// prefix or the interleaved "OFL0X" marker and must upload with
+// "office_sheet_file".
 func TestSheetMediaParentType(t *testing.T) {
 	t.Parallel()
 	cases := []struct {
@@ -40,6 +41,13 @@ func TestSheetMediaParentType(t *testing.T) {
 		{"fake_office token, only the prefix", fakeOfficePrefix, officeSheetFileParentType},
 		{"local_office imported token", "local_office_abc123", officeSheetFileParentType},
 		{"local_office token, only the prefix", localOfficePrefix, officeSheetFileParentType},
+		{"interleaved OFL0X office token", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa", officeSheetFileParentType},
+		{"interleaved exlcn token", "abcdeefghxijkllmnopcqrstnuv", sheetImageParentType},
+		{"interleaved shtcn native token", "abcdsefghhijkltmnopcqrstnuv", sheetImageParentType},
+		{"interleaved pptcn token", "abcdpefghpijkltmnopcqrstnuv", sheetImageParentType},
+		{"interleaved wodcn token", "abcdwefghoijkldmnopcqrstnuv", sheetImageParentType},
+		{"interleaved OFL0X marker with short length", "aaaaOaaaaFaaaaLaaaa0aaaaXaa", sheetImageParentType},
+		{"interleaved OFL0X marker with long length", "aaaaOaaaaFaaaaLaaaa0aaaaXaaaa", sheetImageParentType},
 		{"fake_office prefix mid-string is not matched", "shtfake_office_abc", sheetImageParentType},
 		{"local_office prefix mid-string is not matched", "shtlocal_office_abc", sheetImageParentType},
 	}
@@ -57,7 +65,7 @@ func TestSheetMediaParentType(t *testing.T) {
 // to end (the Execute path the dry-run tests don't reach), asserting the
 // parent_type that actually goes out on the wire is derived from the token: a
 // native spreadsheet uploads as sheet_image, an imported "office" spreadsheet
-// (fake_office_-prefixed token) as office_sheet_file.
+// (legacy prefix or interleaved OFL0X marker) as office_sheet_file.
 func TestUploadSheetImage_ParentType(t *testing.T) {
 	cases := []struct {
 		name           string
@@ -67,6 +75,7 @@ func TestUploadSheetImage_ParentType(t *testing.T) {
 		{"native spreadsheet", "shtcnTOK123", sheetImageParentType},
 		{"fake_office imported spreadsheet", "fake_office_abc123", officeSheetFileParentType},
 		{"local_office imported spreadsheet", "local_office_abc123", officeSheetFileParentType},
+		{"interleaved OFL0X imported spreadsheet", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa", officeSheetFileParentType},
 	}
 	for _, tc := range cases {
 		t.Run(tc.name, func(t *testing.T) {
--- a/skills/lark-calendar/SKILL.md
+++ b/skills/lark-calendar/SKILL.md
@@ -16,14 +16,16 @@ metadata:

 ## 身份

-日程操作默认使用 `--as user`（查看和管理当前用户的日程）。`--as bot` 只能访问 bot 自己的（空）日历，会拿到空结果——不要用 bot 身份查用户日程。
+按**日程归属**选身份：
+
+- 查看/管理登录用户本人的日程 → `--as user`（默认，绝大多数场景）。
+- 查看/管理 bot 自己创建/拥有的日程 → `--as bot` 

 ```bash
-# BAD — bot 身份查用户日程，返回空列表
-lark-cli calendar +agenda --as bot
-
-# GOOD — user 身份查日程
+# 用户本人日程 → user
 lark-cli calendar +agenda --as user
+# bot 自建或参与的日程 → bot
+lark-cli calendar +agenda --as bot
 ```

 ## Shortcuts
@@ -48,7 +50,7 @@ lark-cli calendar +agenda --as user
 lark-cli calendar +get --calendar-id <calendar_id> --event-id <event_id>
 ```

-读取日程时同时返回 `description`（纯文本）和 `description_rich`（**Markdown** 富文本）两个字段：`description` 存纯文本，`description_rich` 存富文本；仅有纯文本描述时，`description_rich` 会用该纯文本兜底填充（两字段值相同）。创建/更新日程时只传 `description_rich`（Markdown）。
+日程描述统一使用 `description` 一个字段，按 **Markdown** 富文本处理。读取日程时 `description` 返回 Markdown 富文本（仅有纯文本描述时返回该纯文本）；创建/更新日程时也通过 `--description` 传入 Markdown。

 ### `+search-event` — 按关键词、时间范围和参会人搜索日程

@@ -188,6 +190,8 @@ lark-cli contact +search-user --query <query> --as user
 lark-cli im +chat-search --query <query> --as user
 ```

+> 搜索用户接口不支持 bot 身份，必须用 `--as user`；搜到的 `ou_` open_id 用于日程参与人操作（如添加日程参与人）。
+
 ## 不在本 skill 范围

 - 查询过去的视频会议记录 → [lark-vc](../lark-vc/SKILL.md)
--- a/skills/lark-calendar/references/lark-calendar-create.md
+++ b/skills/lark-calendar/references/lark-calendar-create.md
@@ -32,20 +32,19 @@ lark-cli calendar +create --summary "..." --start "..." --end "..." \
 | `--summary <text>` | 否 | 日程标题。注意：标题中不应该出现时间、地点、人物信息 |
 | `--start <time>` | 是 | 开始时间（ISO 8601，如 `2026-03-12T14:00+08:00`） |
 | `--end <time>` | 是 | 结束时间（ISO 8601） |
-| `--description-rich <markdown>` | 否 | 日程描述，统一使用此字段，格式为 **Markdown**。提供会议议程、活动内容、注意事项或链接等。支持加粗、斜体、下划线（`<u>...</u>`）、删除线、链接 `[文本](url)`、标题（`# ` 到 `### `，最多三级）、引用（`> `）、有序/无序列表、GFM 表格（`\| 列1 \| 列2 \|` + 分隔行 `\| --- \| --- \|`）、以及图片 `![图片名](图片URL)`（标准 Markdown 图片语法：远程 URL 原样使用；**本地图片路径**（相对路径、且位于当前工作目录内）会自动上传到云盘并在端上内联渲染——绝对路径或工作目录之外的路径会报错；端上已有图片读回为 Markdown 图片）。飞书文档 URL（直接粘贴裸链接，或写成 `[文本](url)`）会自动解析为内联文档，端上展示文档标题而非裸链接。支持 `@文件路径` 或 `-`（stdin）读取。**禁止**用 `***文本***` 同时表示加粗+斜体（端上会残留 `*`）；应嵌套书写，如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。|
-| `--attendee-ids <id_list>` | 否 | 参与人 ID 列表（逗号分隔）。支持用户（`ou_`）、群组（`oc_`）和会议室（`omm_`）。AI 提取时请务必保留对应前缀 |
+| `--description <markdown>` | 否 | 日程描述，统一使用此字段，格式为 **Markdown**。提供会议议程、活动内容、注意事项或链接等。支持加粗、斜体、下划线（`<u>...</u>`）、删除线、链接 `[文本](url)`、标题（`# ` 到 `### `，最多三级）、引用（`> `）、有序/无序列表、GFM 表格（`\| 列1 \| 列2 \|` + 分隔行 `\| --- \| --- \|`）、以及图片 `![图片名](图片URL)`（标准 Markdown 图片语法：远程 URL 原样使用；**本地图片路径**（相对路径、且位于当前工作目录内）会自动上传到云盘并在端上内联渲染——绝对路径或工作目录之外的路径会报错；端上已有图片读回为 Markdown 图片）。飞书文档 URL（直接粘贴裸链接，或写成 `[文本](url)`）会自动解析为内联文档，端上展示文档标题而非裸链接。支持 `@文件路径` 或 `-`（stdin）读取。**禁止**用 `***文本***` 同时表示加粗+斜体（端上会残留 `*`）；应嵌套书写，如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。|
+| `--attendee-ids <id_list>` | 否 | 参与人 ID 列表（逗号分隔）。支持用户（`ou_`）、群组（`oc_`）和会议室（`omm_`）。AI 提取时请务必保留对应前缀。bot 可作为合法参会人，无需剔除 |
 | `--calendar-id <id>` | 否 | 日历 ID（省略则使用主日历） |
 | `--rrule <rrule>` | 否 | 重复日程的重复性规则，规则设置方式参考rfc5545。示例值："FREQ=DAILY;INTERVAL=1;UNTIL=<具体日期>" |
 | `--dry-run` | 否 | 预览 API 调用，不执行 |

 > 当用户表达'每周 X'、'每周重复'、'连续 N 周'时，必须使用 rrule 创建重复性日程，而非创建多个独立日程
-> `--description-rich` 行内同时加粗和斜体时，**禁止**写 `***文本***`（端上会残留 `*`）；必须让 `**` 与 `*` 各自成对嵌套，例如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。
+> `--description` 行内同时加粗和斜体时，**禁止**写 `***文本***`（端上会残留 `*`）；必须让 `**` 与 `*` 各自成对嵌套，例如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。
 > 自动设置 `attendee_ability: "can_modify_event"`，参会人可查看彼此并编辑日程。
 > 自动设置 `free_busy_status: "busy"`，默认日程忙闲状态为忙碌。
 > 自动设置 `reminders: [{"minutes": 5}]`，默认日程开始前 5 分钟提醒。
 > 自动设置 `vchat: {"vc_type": "vc"}`，默认日程包含飞书视频会议。如需其他视频会议类型或不含视频会议，请使用完整 API 命令。
 > 失败保护：若添加参会人失败（如 open_id 错误），CLI 会自动删除刚创建的空日程（回滚，不通知参会人）。
-> 搜索用户接口不支持 bot 身份，需用 `--as user` 进行搜索。
 > 审批会议室：`+create` 不暴露低频字段 `attendees[].approval_reason`。如果会议室要求审批，请使用用户身份先创建日程，再用完整 API `calendar event.attendees create --as user` 添加会议室并传 `approval_reason`。

 ## 高级用法（完整 API 命令）
--- a/skills/lark-calendar/references/lark-calendar-recurring.md
+++ b/skills/lark-calendar/references/lark-calendar-recurring.md
@@ -28,8 +28,8 @@

 | 步骤 | 命令 | 说明 |
 |------|------|------|
-| 1 | `lark-cli calendar +update --event-id <原重复日程ID> --summary ... --description-rich ...` | 更新原重复性日程的标题/描述等 |
-| 2 | `lark-cli calendar +update --event-id <例外ID> --summary ... --description-rich ...` （逐个） | 同步更新例外日程的对应字段 |
+| 1 | `lark-cli calendar +update --event-id <原重复日程ID> --summary ... --description ...` | 更新原重复性日程的标题/描述等 |
+| 2 | `lark-cli calendar +update --event-id <例外ID> --summary ... --description ...` （逐个） | 同步更新例外日程的对应字段 |

 > 理由：例外已脱离原重复性日程独立存在，不会自动继承原日程的更新。

--- a/skills/lark-calendar/references/lark-calendar-room-find.md
+++ b/skills/lark-calendar/references/lark-calendar-room-find.md
@@ -50,12 +50,13 @@ lark-cli calendar +room-find \
 | `--room-name <text>` | 否 | 会议室名称约束，支持以**英文逗号**分隔传入多个名称。仅当用户明确提到会议室专名、会议室号或编号区间时使用。 |
 | `--min-capacity <n>` | 否 | 会议室最小容纳人数。当用户明确参会人数或提出“至少容纳N人”等要求时，提取数字放入此参数，必须为正整数。 |
 | `--max-capacity <n>` | 否 | 会议室最大容纳人数。用于过滤过大空间，必须为正整数。 |
-| `--attendee-ids <id_list>` | 否 | 参会对象 ID 列表。支持用户 ID（`ou_` 前缀）和群组 ID（`oc_` 前缀），多个 ID 以逗号分隔。 |
+| `--attendee-ids <id_list>` | 否 | 参会对象 ID 列表。支持用户 ID（`ou_` 前缀）和群组 ID（`oc_` 前缀），多个 ID 以逗号分隔。**不要传入 bot 的 open_id**：bot 是虚拟身份，不占会议室席位、无会议室偏好，传入只会干扰推荐结果。 |
 | `--event-rrule <rrule>` | 否 | 重复日程的重复性规则，规则设置方式参考rfc5545。**【⚠️注意：系统绝对不支持 COUNT，如需限制重复次数，必须转为 UNTIL】**。示例值："FREQ=DAILY;INTERVAL=1" |
 | `--timezone <tz>` | 否 | 对话中明确提及的预约日程所使用的时区（默认取用户设备时区，例如 `Asia/Shanghai`） |

 ## 规则

+- 构造 `--attendee-ids` 前，先剔除 bot 参会人：bot 不占席位、无偏好，不应参与会议室推荐。
 - 多个 `--slot` 会由 CLI 内部并发调用单时间块接口，再聚合成一次输出
 - `+room-find` 的时间输入必须是**确定时间块**，不是时间区间搜索。
 - 如果是重复性日程，必须校验返回中的 `reserve_until_time`（该会议室最晚可预约时间）是否覆盖 `event-rrule` 对应的重复范围。
--- a/skills/lark-calendar/references/lark-calendar-schedule-clear-time.md
+++ b/skills/lark-calendar/references/lark-calendar-schedule-clear-time.md
@@ -39,6 +39,7 @@ lark-cli calendar +freebusy --start "<start>" --end "<end>"
 ```

 规则：
+- 参与人含 **bot**：无需为 bot 查询忙闲。bot 是虚拟身份，可并行多个会议、无忙闲语义，检查它没有意义。
 - 参与人过多（超过 5 人）：仅查询**当前用户**及少数核心人员忙闲即可
 - 参与人含**群组**：无需展开群组成员查询忙闲
 - 如果用户是从 `+suggestion` 确认了时间块后进入本分支的，**无需再调用 `+freebusy`**
--- a/skills/lark-calendar/references/lark-calendar-suggestion.md
+++ b/skills/lark-calendar/references/lark-calendar-suggestion.md
@@ -45,7 +45,7 @@ lark-cli calendar +suggestion \
 | ------------------------------- | ----- | ------------------------------------------------------------------- |
 | `--start <time>`                | 否     | 搜索区间开始时间（支持日期/ISO 8601等格式，默认**当前时间**）                                |
 | `--end <time>`                  | 否     | 搜索区间结束时间（默认与 `--start` 属于同一天，自动取当天结束时间）                                                     |
-| `--attendee-ids <id_list>`     | 否     | 目标参与人 ID 列表。提取对应实体的 ID。支持用户（`ou_` 前缀）和群组（`oc_` 前缀）。多个 ID 使用英文逗号分隔 |
+| `--attendee-ids <id_list>`     | 否     | 目标参与人 ID 列表。提取对应实体的 ID。支持用户（`ou_` 前缀）和群组（`oc_` 前缀）。多个 ID 使用英文逗号分隔。**不要传入 bot 的 open_id**：bot 是虚拟身份，可并行多个会议、无忙闲语义，传入会干扰推荐时段的忙闲计算。 |
 | `--event-rrule <rrule>`         | 否     | 重复日程的重复性规则，规则设置方式参考rfc5545。**【⚠️注意：系统绝对不支持 COUNT，如需限制重复次数，必须转为 UNTIL】**。示例值："FREQ=DAILY;INTERVAL=1"                                              |
 | `--duration-minutes <min>`      | 否     | 会议时长（分钟）。优先使用用户显式指定的值，若未指定则尝试根据上下文推断，推断失败则不传                                        |
 | `--timezone <tz>`               | 否     | 对话中明确提及的预约日程所使用的时区（默认取用户设备时区，例如 `Asia/Shanghai`）                                |
--- a/skills/lark-calendar/references/lark-calendar-update.md
+++ b/skills/lark-calendar/references/lark-calendar-update.md
@@ -12,7 +12,7 @@
 lark-cli calendar +update \
  --event-id "<EVENT_ID>" \
  --summary "产品评审" \
-  --description-rich "评审需求范围、排期与风险" \
+  --description "评审需求范围、排期与风险" \
  --start "2026-03-12T14:00+08:00" \
  --end "2026-03-12T15:00+08:00"

@@ -43,7 +43,7 @@ lark-cli calendar +update \
 | `--event-id <id>` | 是 | 要更新的日程 ID。重复性日程请根据操作范围选择 ID，详见 [重复性日程操作规范](lark-calendar-recurring.md) |
 | `--calendar-id <id>` | 否 | 日历 ID（省略则使用 `primary`） |
 | `--summary <text>` | 否 | 新日程标题。仅在显式传入 `--summary` 时更新；若传空字符串，会把标题清空 |
-| `--description-rich <markdown>` | 否 | 新日程描述，统一使用此字段，格式为 **Markdown**（加粗、斜体、下划线 `<u>...</u>`、删除线、链接 `[文本](url)`、标题 `# `~`### `（最多三级）、引用 `> `、有序/无序列表、GFM 表格 `\| 列1 \| 列2 \|` + 分隔行 `\| --- \| --- \|`、以及图片 `![图片名](图片URL)`（标准 Markdown 图片语法：远程 URL 原样使用；**本地图片路径**（相对路径、且位于当前工作目录内）会自动上传到云盘并在端上内联渲染——绝对路径或工作目录之外的路径会报错；端上已有图片读回为 Markdown 图片）。飞书文档 URL（裸链接或 `[文本](url)`）会自动解析为内联文档，端上展示文档标题。支持 `@文件路径` 或 `-`（stdin）读取。仅在显式传入时更新；传空字符串 `""` 会清空描述。**禁止**用 `***文本***` 同时表示加粗+斜体（端上会残留 `*`）；应嵌套书写，如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。 |
+| `--description <markdown>` | 否 | 新日程描述，统一使用此字段，格式为 **Markdown**（加粗、斜体、下划线 `<u>...</u>`、删除线、链接 `[文本](url)`、标题 `# `~`### `（最多三级）、引用 `> `、有序/无序列表、GFM 表格 `\| 列1 \| 列2 \|` + 分隔行 `\| --- \| --- \|`、以及图片 `![图片名](图片URL)`（标准 Markdown 图片语法：远程 URL 原样使用；**本地图片路径**（相对路径、且位于当前工作目录内）会自动上传到云盘并在端上内联渲染——绝对路径或工作目录之外的路径会报错；端上已有图片读回为 Markdown 图片）。飞书文档 URL（裸链接或 `[文本](url)`）会自动解析为内联文档，端上展示文档标题。支持 `@文件路径` 或 `-`（stdin）读取。仅在显式传入时更新；传空字符串 `""` 会清空描述。**禁止**用 `***文本***` 同时表示加粗+斜体（端上会残留 `*`）；应嵌套书写，如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。 |
 | `--start <time>` | 否 | 新开始时间（ISO 8601，如 `2026-03-12T14:00+08:00`）。更新日程时间时必须同时传 `--end` |
 | `--end <time>` | 否 | 新结束时间（ISO 8601）。更新日程时间时必须同时传 `--start` |
 | `--rrule <rrule>` | 否 | 新重复规则（RFC5545）。**不要使用 COUNT；如需限制次数，推算后转为 UNTIL** |
@@ -52,17 +52,18 @@ lark-cli calendar +update \
 | `--notify` | 否 | 是否发送更新通知，默认 `true`。可用 `--notify=false` 静默更新 |
 | `--dry-run` | 否 | 预览 API 调用，不执行 |

-至少需要提供一个动作：`--summary`、`--description-rich`、`--start/--end`、`--rrule`、`--add-attendee-ids` 或 `--remove-attendee-ids`。
+至少需要提供一个动作：`--summary`、`--description`、`--start/--end`、`--rrule`、`--add-attendee-ids` 或 `--remove-attendee-ids`。

 ## 使用规则

 - `--add-attendee-ids` 是**增量添加**，不是替换最终参与人列表。不要用它表达“只保留这些人”。
- 对 `--summary`、`--description-rich`，CLI 以“是否显式传入该 flag”判断是否更新，而不是以“值是否为空”判断；如果显式传入空字符串，会把对应字段清空。
- 日程描述统一走 `--description-rich`（Markdown）。`--description`（纯文本）已废弃并从帮助中隐藏。
+- 对 `--summary`、`--description`，CLI 以“是否显式传入该 flag”判断是否更新，而不是以“值是否为空”判断；如果显式传入空字符串，会把对应字段清空。
+- 日程描述统一走 `--description`（按 Markdown 富文本处理）。
 - 行内同时加粗和斜体时，**禁止**写 `***文本***`（端上会残留 `*`）；必须让 `**` 与 `*` 各自成对嵌套，例如 `**<u>*~~文本~~*</u>**` 或 `*<u>**~~文本~~**</u>*`。
 - 只想增删参会人或会议室时，不需要同时传 `--summary`、`--start`、`--end` 等日程字段。
 - 只想修改标题、描述、时间或重复规则时，不需要同时传 `--add-attendee-ids` 或 `--remove-attendee-ids`。
 - 如需替换某个参与人、群组或会议室，使用 `--remove-attendee-ids <旧ID>` + `--add-attendee-ids <新ID>`。
+- bot 可作为合法参会人添加，无需剔除。
 - 会议室是 resource attendee，必须使用 `omm_` ID 添加到参会人列表，不能脱离日程单独预定。
 - 更新重复性日程时，必须先确定操作范围（仅此次/全部/此次及后续），然后按 [重复性日程操作规范](lark-calendar-recurring.md) 执行。
 - 当同一次命令组合多个动作时，执行顺序为“日程字段 -> 移除参会人 -> 添加参会人”。若中途失败，不会自动回滚已成功步骤；错误信息会说明已完成的步骤。
--- a/skills/lark-drive/references/lark-drive-search.md
+++ b/skills/lark-drive/references/lark-drive-search.md
@@ -26,11 +26,16 @@
 > **`--query` 最长 30 个字符**：按字符数（Unicode 码点）算，中文每字算 1 个，与 ASCII 同口径；超过 30 会被服务端拒绝（`99992402 field validation failed`，**是报错不是截断**）。长关键词必须先压缩成核心实体 + 主题词（如把整句问题压成「项目名 + 主题」再搜），不要把整句原问塞进 `--query`。
 >
 > **列表型请求不要硬塞关键词**：如果用户只是要求"我这月创建的所有文档"、"最近半年我编辑过的文档"、"按类型分类统计"这类范围浏览 / 汇总请求，且没有给出标题片段或业务关键词，应使用 `--query ""` 搭配 `--created-by-me`、`--mine`、`--created-*`、`--edited-*`、`--doc-types` 等过滤条件。不要把"查找"、"所有文档"、"最近更新过"、"按类型分类统计"这类动作词或统计意图放进 `--query`，否则会把本来应靠 filter 命中的结果过度收窄。
+>
+> **标题词 + 正文词联合搜索**：如果用户同时给出标题关键词和正文关键词，并要求同一资源同时满足两项条件，优先执行一条普通联合搜索：`lark-cli drive +search --query "标题词 正文词"`，并在同一条命令中叠加用户指定的 `--folder-tokens`、`--doc-types` 等过滤条件。不要把这种联合搜索拆成“标题搜索 + 正文搜索”后自行拼交集；也不要把 `--only-title` 或 `intitle:` 用作主候选路径。只有用户明确只查标题时，才使用 `--only-title` 或 `intitle:`。
+>
+> 用户要求最终返回 N 条时，N 是输出上限，不等于 `--page-size N`。逐页根据 `title` 和 `summary_highlighted` 保留同时满足两项条件的候选；有效候选不足 N 且 `has_more=true` 时，保持同一 query 和过滤条件，使用 `--page-token` 继续，最多检查 3 页。摘要不足以判断正文条件时，只对标题已匹配的候选串行读取正文，确认一个再处理下一个，找到 N 条后停止；不要并发拉取正文。检查 3 页后仍不足时，返回已确认结果并建议用户调整标题词、正文词或搜索范围，不要无界扫描。

 ### 自然语言 → 命令映射速查

 | 用户说 | 命令 |
 |---|---|
+| 标题含某词且正文含某词，限定文件夹内最多 N 个结果（N 为最终输出上限；按上文规则分页筛选，勿作为 `--page-size`） | `lark-cli drive +search --query "标题词 正文词" --folder-tokens <FOLDER_TOKEN>` |
 | 我这月创建的所有文档，按类型分类统计 | `lark-cli drive +search --query "" --created-by-me --created-since "<YYYY-MM-DD>" --created-until "<YYYY-MM-DD>"` |
 | 最近半年我编辑过的文档，看看哪些最近更新过 | `lark-cli drive +search --query "" --edited-since 6m --sort edit_time` |
 | 最近一个月我编辑过的文档 | `lark-cli drive +search --query "" --edited-since 1m` |
@@ -217,7 +222,7 @@ stdout 的 JSON 输出不受影响。`open_time` / `create_time` 不做 snap。
  - **日历表达**（"上个月"、"上周"、"本月"、"前年"、"今年 3 月"等明确日历单位）→ **必须算出绝对 `YYYY-MM-DD` 边界**（如"上个月" = 上一个日历月的 1 号 → 当月 1 号），**不要近似成 `1m`/`2m`**：CLI 里 `m` 是固定 30 天、`y` 固定 365 天，跟日历差 0-3 天，月末月初尤其容易偏出去
  - 文档中的 `"<YYYY-MM-DD>"` 是运行时占位符：执行命令前按当前日期计算并替换。例如"本月"应替换为本月第一天和下月第一天，不要把示例生成时的月份硬编码进答案
  - 绝对日期 → 直接 `YYYY-MM-DD` 或 RFC3339
- **分页策略**：默认只返回第一页，并说明 `has_more` 和下一页命令。只有用户明确要"全部 / 全量 / 继续翻"才继续。单轮翻页上限 5 页。
+- **分页策略**：默认只返回第一页，并说明 `has_more` 和下一页命令。用户明确要"全部 / 全量 / 继续翻"时继续；标题词 + 正文词联合搜索尚未找到足够的有效 Top N 候选时，按上文规则最多检查 3 页。其他场景单轮翻页上限 5 页。
 - **原始返回**：用户要求"原始数据"、"接口返回"时用 `--format json`，不做客户端精确过滤或摘要重写。

 ## 权限
--- a/skills/lark-slides/SKILL.md
+++ b/skills/lark-slides/SKILL.md
@@ -141,7 +141,7 @@ lark-cli auth login --domain slides
 - [asset-planning.md](references/asset-planning.md)（新建 / 大幅改写）
 - [validation-checklist.md](references/validation-checklist.md)（创建 / 大幅改写后）

-按需再读：
+调用相关命令前必须读取相关的文档以了解命令的使用方式：

 - 创建：[`lark-slides-create.md`](references/lark-slides-create.md)
 - 阅读：[`lark-slides-xml-presentations-get.md`](references/lark-slides-xml-presentations-get.md)
@@ -223,7 +223,7 @@ Step 3: 按 slide_plan.json 生成 XML → 创建
  - 创建方式按“创建方式选择”判断；图片、复杂 XML、转义和 3350001 排查按 lark-slides-create.md、media-upload.md、troubleshooting.md 执行

 Step 4: 审查 & 交付
-  - 创建完成后，必须用 `slides +xml-get` 读取全文 XML，并按 validation-checklist.md 做显式验证记录，包括 XML 文本重叠检查
+  - 创建完成后，必须用 `slides +xml-get --presentation <xml_presentation_id>` 读取全文 XML，并按 validation-checklist.md 做显式验证记录，包括 XML 文本重叠检查
  - 失败或部分成功按 troubleshooting.md 处理；局部问题优先用 `+replace-slide` 修正
  - 没问题 → 交付：使用 NotifyHuman 工具交付 PPT 链接
 ```
@@ -314,7 +314,8 @@ Shortcut 是对常用操作的高级封装（`lark-cli slides +<verb> [flags]`
 | 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） |
-| [`+xml-get`](references/lark-slides-xml-presentations-get.md) | 读取全文 XML 并保存到本地文件，避免终端输出被截断 |
+| [`+xml-get`](references/lark-slides-xml-presentations-get.md) | 读取全文 XML，用 `--presentation` 指定演示文稿的 `xml_presentation_id`，用 `--output` 把 XML 存到本地文件（必须是 CWD 内的相对路径，如 `.lark-slides/plan/<deck>/readback.xml`） |
+| [`+screenshot`](references/lark-slides-screenshot.md) | 把幻灯片页面截图保存为本地图片，用 `--slide-number` 指定页号（从 1 开始，多页重复传入，一次最多 10 页），用 `--output-dir` 指定保存目录（必须是 CWD 内的相对路径，默认 `.lark-slides/screenshots`），失败时降级到 XML 回读等非截图检查 |
 | [`+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 |
 | [`+replace-pages`](references/lark-slides-replace-pages.md) | 在原演示文稿内批量重建多个页面：先创建新页到旧页前，再删除旧页；适合已有 Slides 的多页大改，不新建链接 |
--- a/skills/lark-slides/references/slides_xml_schema_definition.xml
+++ b/skills/lark-slides/references/slides_xml_schema_definition.xml
@@ -194,14 +194,13 @@
    <xs:simpleType name="FontSizeType">
        <xs:annotation>
            <xs:documentation>
-                字体大小, 使用正整数, 单位px
-                示例：12, 14, 16, 18, 20, 24, 28, 32 等
+                字体大小, 浮点数, 范围 [1, 4000], 单位px
+                示例：10, 10.5, 12, 14, 16, 18, 20, 24, 28, 32 等
            </xs:documentation>
        </xs:annotation>
-        <xs:restriction base="xs:positiveInteger">
-            <xs:minInclusive value="6"/>
-            <xs:maxInclusive value="400"/>
-            <xs:pattern value="[0-9]+"/>
+        <xs:restriction base="xs:double">
+            <xs:minInclusive value="1"/>
+            <xs:maxInclusive value="4000"/>
        </xs:restriction>
    </xs:simpleType>

@@ -211,6 +210,52 @@
        </xs:restriction>
    </xs:simpleType>

+    <xs:simpleType name="AutoStartAtType">
+        <xs:annotation>
+            <xs:documentation>
+                有序列表起始编号, 取值范围 [1, 32767]
+            </xs:documentation>
+        </xs:annotation>
+        <xs:restriction base="xs:positiveInteger">
+            <xs:minInclusive value="1"/>
+            <xs:maxInclusive value="32767"/>
+        </xs:restriction>
+    </xs:simpleType>
+
+    <xs:simpleType name="BulletSizeType">
+        <xs:annotation>
+            <xs:documentation>
+                列表符号大小, 二选一:
+                - 百分比字符串（相对于文本字号）, 取值范围 25%-400%, 如 "100%"
+                - 绝对像素值, 取值范围 6-400, 如 "14"
+            </xs:documentation>
+        </xs:annotation>
+        <xs:union>
+            <xs:simpleType>
+                <xs:restriction base="xs:string">
+                    <xs:pattern value="(2[5-9]|[3-9][0-9]|[1-3][0-9]{2}|400)%"/>
+                </xs:restriction>
+            </xs:simpleType>
+            <xs:simpleType>
+                <xs:restriction base="xs:string">
+                    <xs:pattern value="[6-9]|[1-9][0-9]|[1-3][0-9]{2}|400"/>
+                </xs:restriction>
+            </xs:simpleType>
+        </xs:union>
+    </xs:simpleType>
+
+    <xs:simpleType name="BulletCharType">
+        <xs:annotation>
+            <xs:documentation>
+                自定义列表符号, 如 "★", "→", "✓", "◆", 也支持 emoji
+            </xs:documentation>
+        </xs:annotation>
+        <xs:restriction base="xs:string">
+            <xs:minLength value="1"/>
+            <xs:maxLength value="8"/>
+        </xs:restriction>
+    </xs:simpleType>
+
    <!-- 文本类型枚举 -->
    <xs:simpleType name="TextType">
        <xs:annotation>
@@ -232,6 +277,35 @@
        </xs:restriction>
    </xs:simpleType>

+    <!-- 动态文本字段类型枚举 -->
+    <xs:simpleType name="FieldType">
+        <xs:annotation>
+            <xs:documentation>
+                动态文本字段类型:
+                - slidenum: 当前幻灯片页码
+                - datetime: 默认日期时间格式
+                - datetime1-datetime13: 预定义日期时间格式
+            </xs:documentation>
+        </xs:annotation>
+        <xs:restriction base="xs:string">
+            <xs:enumeration value="slidenum"><xs:annotation><xs:documentation>当前幻灯片页码</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime"><xs:annotation><xs:documentation>浏览器默认日期格式, 例如 2026/7/14</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime1"><xs:annotation><xs:documentation>日期格式 M/D/YYYY, 例如 10/12/2007</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime2"><xs:annotation><xs:documentation>日期格式 dddd, MMMM D, YYYY, 例如 Friday, October 12, 2007</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime3"><xs:annotation><xs:documentation>日期格式 D MMMM YYYY, 例如 12 October 2007</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime4"><xs:annotation><xs:documentation>日期格式 MMMM D, YYYY, 例如 October 12, 2007</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime5"><xs:annotation><xs:documentation>日期格式 D-MMM-YY, 例如 12-Oct-07</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime6"><xs:annotation><xs:documentation>日期格式 MMMM YY, 例如 October 07</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime7"><xs:annotation><xs:documentation>日期格式 MMM-YY, 例如 Oct-07</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime8"><xs:annotation><xs:documentation>日期时间格式 M/D/YYYY h:mm A, 例如 10/12/2007 4:28 PM</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime9"><xs:annotation><xs:documentation>日期时间格式 M/D/YYYY h:mm:ss A, 例如 10/12/2007 4:28:34 PM</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime10"><xs:annotation><xs:documentation>时间格式 HH:mm, 例如 16:28</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime11"><xs:annotation><xs:documentation>时间格式 HH:mm:ss, 例如 16:28:34</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime12"><xs:annotation><xs:documentation>时间格式 h:mm A, 例如 4:28 PM</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="datetime13"><xs:annotation><xs:documentation>时间格式 h:mm:ss A, 例如 4:28:34 PM</xs:documentation></xs:annotation></xs:enumeration>
+        </xs:restriction>
+    </xs:simpleType>
+
    <!-- 文本对齐 -->
    <xs:simpleType name="TextAlignType">
        <xs:restriction base="xs:string">
@@ -781,21 +855,64 @@
        <xs:attribute name="heightScale" type="sml:ArrowScaleType" use="optional"/>
    </xs:complexType>

+    <!-- 裁剪方位枚举类型 -->
+    <xs:simpleType name="CropAnchorType">
+        <xs:annotation>
+            <xs:documentation>
+                裁剪方位枚举, 用于指定保留原图的哪个区域
+
+                - top: 保留顶部, 裁掉底部多余部分
+                - bottom: 保留底部, 裁掉顶部多余部分
+                - left: 保留左侧, 裁掉右侧多余部分
+                - right: 保留右侧, 裁掉左侧多余部分
+
+                居中场景不需要设置 anchor, 不设置 offset 即为默认居中裁剪
+            </xs:documentation>
+        </xs:annotation>
+        <xs:restriction base="xs:string">
+            <xs:enumeration value="top"><xs:annotation><xs:documentation>保留顶部, 裁掉底部多余部分</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="bottom"><xs:annotation><xs:documentation>保留底部, 裁掉顶部多余部分</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="left"><xs:annotation><xs:documentation>保留左侧, 裁掉右侧多余部分</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="right"><xs:annotation><xs:documentation>保留右侧, 裁掉左侧多余部分</xs:documentation></xs:annotation></xs:enumeration>
+        </xs:restriction>
+    </xs:simpleType>
+
    <!-- 裁剪类型定义 -->
    <xs:complexType name="CropType">
        <xs:annotation>
            <xs:documentation>
-                裁剪配置: 原图填充到预裁剪区域，再根据offset裁出最终尺寸
+                裁剪配置: 将原图裁剪到目标尺寸 (img 元素的 width × height)

                可选属性:
-                type: 裁剪形状，默认rect
-                leftOffset, rightOffset, topOffset, bottomOffset: 边缘偏移量(px)。正值向内裁剪，负值向外扩展留白，0值对齐边缘
-                presetHandlers: 控制点配置，对应ECMA预设形状的控制点。单个或多个数字，多个用逗号分隔。示例: type="rect"且presetHandlers="60"时为圆角矩形，圆角半径60px
+                type: 裁剪形状, 默认 rect
+                anchor: 裁剪方位 (top/bottom/left/right), 参见 CropAnchorType
+                leftOffset / rightOffset / topOffset / bottomOffset: 四向偏移量 (px), 正值向内裁剪、负值向外扩展留白、0对齐边缘
+                presetHandlers: 控制点配置, 对应 ECMA 预设形状的控制点。单个或多个数字, 多个用逗号分隔。
+                示例: type="rect" 且 presetHandlers="60" 时为圆角矩形, 圆角半径 60px

-                说明: 指定offset时，若预裁剪尺寸与原图比例不一致会产生拉伸变形。无法确定原图比例时，不要指定offset
+                【推荐用法】使用 anchor 指定裁剪方位:
+                - 不设置 anchor 时: 默认按图片居中裁剪
+                - 设置 anchor 时: 按指定方位裁剪, 例如 anchor="top" 表示保留顶部、裁掉底部多余部分
+                - 使用 anchor 后, 不需要再设置 offset
+                - anchor 模式下原图按等比缩放后裁剪, 不会发生拉伸或压缩
+
+                【进阶用法】使用 offset 精细控制裁剪边界:
+                - 适用于用户在编辑器中手动调整裁剪、或从外部协议导入的场景
+                - 原图先填充到预裁剪区域, 再根据 offset 从四边裁出最终尺寸
+                - 注意: 若预裁剪尺寸与原图比例不一致会产生拉伸变形; 无法确定原图比例时, 不要指定 offset
+
+                【优先级】
+                如果同时设置了 anchor 和 offset, 以 anchor 为准, offset 被忽略
+
+                典型用法:
+                <crop/>                                      居中裁剪 (默认行为)
+                <crop anchor="top"/>                        保留顶部
+                <crop anchor="left"/>                       保留左侧
+                <crop type="rect" presetHandlers="60"/>     圆角矩形裁剪, 默认居中
            </xs:documentation>
        </xs:annotation>
        <xs:attribute name="type" type="sml:ShapeType" use="optional" default="rect"/>
+        <xs:attribute name="anchor" type="sml:CropAnchorType" use="optional"/>
        <xs:attribute name="leftOffset" type="xs:double" use="optional"/>
        <xs:attribute name="rightOffset" type="xs:double" use="optional"/>
        <xs:attribute name="topOffset" type="xs:double" use="optional"/>
@@ -959,6 +1076,7 @@
                        <xs:element ref="sml:img"/>
                        <xs:element ref="sml:table"/>
                        <xs:element ref="sml:icon"/>
+                        <xs:element ref="sml:embed"/>
                        <xs:element ref="sml:chart"/>
                        <xs:element ref="sml:undefined"/>
                    </xs:choice>
@@ -1042,6 +1160,10 @@
                - underline: content 级别是否下划线
                - list: content 级别列表类型 bullet/number
                - listStyle: content 级别列表样式
+                - bulletColor: 列表符号颜色（纯色）, 可选
+                - bulletSize: 列表符号大小, 可选, 百分比字符串（相对于文本字号, 取值范围 25%-400%）如 "100%", 或绝对像素值（取值范围 6-400）如 "14"
+                - autoStartAt: 有序列表起始编号, 可选, 取值范围 [1, 32767]; 作为后代段落的初始计数器, 子元素 &lt;p&gt;/&lt;ol&gt; 可通过自身 autoStartAt 重置
+                - bulletChar: 自定义列表符号字符, 可选, 如 "★", "→" 等, 设置后覆盖 listStyle 的符号
                - anchorCenter: 控制文本对齐方式, 优先级高于 textAlign
                - autoFit: 控制文本编辑溢出时处理策略
                - baseline: 上标/下标, 相较于文本基线的偏移量
@@ -1049,6 +1171,13 @@

                注意：如果content子元素不指定属性, 默认继承content的属性值, 如果局部子元素指定了属性, 则使用局部属性值

+                autoStartAt 运行计数器示例（显式指定重置, 未指定沿用前序计数器）:
+                &lt;content autoStartAt="5"&gt;
+                &lt;p list="number"&gt;A&lt;/p&gt;                    &lt;!-- A=5, 继承 content 初始值 --&gt;
+                &lt;p list="number" autoStartAt="10"&gt;B&lt;/p&gt;    &lt;!-- B=10, 本段显式重置 --&gt;
+                &lt;p list="number"&gt;C&lt;/p&gt;                    &lt;!-- C=11, 沿用前序计数器递增 --&gt;
+                &lt;/content&gt;
+
                子元素:
                - p: 段落元素
                - ul: 无序列表元素
@@ -1085,6 +1214,10 @@
            <xs:attribute name="underline" type="xs:boolean" />
            <xs:attribute name="list" type="sml:ListType" default="none"/>
            <xs:attribute name="listStyle" type="sml:ListStyleType" />
+            <xs:attribute name="bulletColor" type="sml:SolidColor" use="optional"/>
+            <xs:attribute name="bulletSize" type="sml:BulletSizeType" use="optional"/>
+            <xs:attribute name="autoStartAt" type="sml:AutoStartAtType" use="optional"/>
+            <xs:attribute name="bulletChar" type="sml:BulletCharType" use="optional"/>
            <xs:attribute name="anchorCenter" type="xs:boolean" default="false" /> <!-- 控制竖排文字是否在垂直方向保持居中 -->
            <xs:attribute name="autoFit" type="sml:AutoFitType" default="no-auto-fit" />
            <xs:attribute name="wrap" type="xs:boolean" default="true" />
@@ -1096,9 +1229,9 @@
        <xs:annotation>
            <xs:documentation>
                段落容器, 支持富文本内容
-                可包含纯文本和内联格式元素(br/strong/em/u/span/del/a/shadow/outline)
+                可包含纯文本和内联格式元素(br/strong/em/u/span/del/a/shadow/outline/formula/field)
                内联元素嵌套：所有内联元素均可包含纯文本或其他内联元素，以实现复杂的格式组合
-                元素自嵌套：除a元素外，其余内联元素支持自身嵌套，当shadow和outline自嵌套时，渲染效果遵循就近原则，以内层定义的样式为准
+                元素自嵌套：除a/formula元素外，其余内联元素支持自身嵌套，当shadow和outline自嵌套时，渲染效果遵循就近原则，以内层定义的样式为准

                空格处理规则：
                - 文本内的连续空格会被合并为单个空格
@@ -1119,6 +1252,8 @@
                - a: 超链接
                - shadow: 文本阴影
                - outline: 文本轮廓
+                - formula: 科学公式（支持数学、物理等）
+                - field: 动态文本字段，元素内容作为不支持动态字段时的降级文本
                属性说明:
                - textAlign: 文本对齐方式
                - lineSpacing: 行间距
@@ -1127,6 +1262,10 @@
                - level: 段落级别, 取值范围 [1,10]
                - list: 列表类型(bullet/number)
                - listStyle: 列表样式
+                - bulletColor: 列表符号颜色（纯色）, 可选
+                - bulletSize: 列表符号大小, 可选, 百分比字符串（相对于文本字号, 取值范围 25%-400%）或绝对像素值（取值范围 6-400）
+                - autoStartAt: 有序列表起始编号, 可选, 取值范围 [1, 32767]; 显式指定时从当前段落起重置计数器, 未指定时沿用同一 content 内的前序计数器
+                - bulletChar: 自定义列表符号字符, 可选
                - marginLeft: 段落左侧缩进宽度
                - indent: 首行缩进宽度
            </xs:documentation>
@@ -1134,6 +1273,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1142,6 +1282,7 @@
                <xs:element ref="sml:a"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
            <xs:attribute name="textAlign" type="sml:TextAlignType" />
            <xs:attribute name="lineSpacing" type="sml:LineSpacingType" default="multiple:1.5"/>
@@ -1151,6 +1292,10 @@
            <xs:attribute name="level" type="sml:LevelType" default="1"/>
            <xs:attribute name="list" type="sml:ListType" default="none"/>
            <xs:attribute name="listStyle" type="sml:ListStyleType"/>
+            <xs:attribute name="bulletColor" type="sml:SolidColor" use="optional"/>
+            <xs:attribute name="bulletSize" type="sml:BulletSizeType" use="optional"/>
+            <xs:attribute name="autoStartAt" type="sml:AutoStartAtType" use="optional"/>
+            <xs:attribute name="bulletChar" type="sml:BulletCharType" use="optional"/>
            <xs:attribute name="marginLeft" type="xs:double" use="optional" />
            <xs:attribute name="indent" type="sml:NonNegativeDouble" use="optional"/>
        </xs:complexType>
@@ -1160,7 +1305,14 @@
    <!-- 无序列表 -->
    <xs:element name="ul">
        <xs:annotation>
-            <xs:documentation>无序列表</xs:documentation>
+            <xs:documentation>
+                无序列表
+                属性说明:
+                - listStyle: 列表样式
+                - bulletColor: 列表符号颜色（纯色）, 可选
+                - bulletSize: 列表符号大小, 可选, 百分比字符串（相对于文本字号, 取值范围 25%-400%）或绝对像素值（取值范围 6-400）
+                - bulletChar: 自定义列表符号字符, 可选, 设置后覆盖 listStyle 的符号
+            </xs:documentation>
        </xs:annotation>
        <xs:complexType>
            <xs:sequence>
@@ -1173,13 +1325,23 @@
                </xs:element>
            </xs:sequence>
            <xs:attribute name="listStyle" type="sml:UnorderedListStyle" default="circle-hollow-square"/>
+            <xs:attribute name="bulletColor" type="sml:SolidColor" use="optional"/>
+            <xs:attribute name="bulletSize" type="sml:BulletSizeType" use="optional"/>
+            <xs:attribute name="bulletChar" type="sml:BulletCharType" use="optional"/>
        </xs:complexType>
    </xs:element>

    <!-- 有序列表 -->
    <xs:element name="ol">
        <xs:annotation>
-            <xs:documentation>有序列表, 可指定序号</xs:documentation>
+            <xs:documentation>
+                有序列表, 可指定序号
+                属性说明:
+                - listStyle: 列表样式
+                - bulletColor: 列表符号颜色（纯色）, 可选
+                - bulletSize: 列表符号大小, 可选, 百分比字符串（相对于文本字号, 取值范围 25%-400%）或绝对像素值（取值范围 6-400）
+                - autoStartAt: 有序列表起始编号, 可选, 取值范围 [1, 32767]; 作用于本列表组的计数器初始值, 子元素 &lt;li@index&gt; 可覆盖单项编号
+            </xs:documentation>
        </xs:annotation>
        <xs:complexType>
            <xs:sequence>
@@ -1193,6 +1355,9 @@
                </xs:element>
            </xs:sequence>
            <xs:attribute name="listStyle" type="sml:OrderedListStyle" default="number-lower-alpha-lower-roman"/>
+            <xs:attribute name="bulletColor" type="sml:SolidColor" use="optional"/>
+            <xs:attribute name="bulletSize" type="sml:BulletSizeType" use="optional"/>
+            <xs:attribute name="autoStartAt" type="sml:AutoStartAtType" use="optional"/>
        </xs:complexType>
    </xs:element>

@@ -1383,7 +1548,7 @@
                alpha: 不透明度[0, 1]

                可选子元素:
-                crop: 裁剪。无标签或所有offset未设置时从左上角自适应裁到width×height
+                crop: 裁剪。无标签 / 空标签 / 仅设 anchor 时按等比缩放后裁剪到 width×height; anchor 指定保留方位 (top/bottom/left/right), 不设 anchor 即居中裁剪; offset 用于精细控制
                reflection: 倒影。无标签代表无倒影,空标签代表使用默认样式
                shadow: 阴影。无标签代表无阴影,空标签代表使用默认样式
                border: 边框。无标签代表无边框,空标签代表使用默认样式(颜色: rgba(43, 47, 54, 1), 宽度: 2)
@@ -1458,6 +1623,49 @@
        </xs:complexType>
    </xs:element>

+    <!-- 嵌入内容元素 -->
+    <xs:element name="embed">
+        <xs:annotation>
+            <xs:documentation>
+                嵌入内容容器, 外层承载 Slides 摆放和效果属性, 内层承载外部标准内容。
+                当前内层内容为标准 SVG, 必须使用 http://www.w3.org/2000/svg 命名空间。
+                后续可扩展承载 HTML 等其他外部内容。
+
+                必需属性:
+                - topLeftX/topLeftY: 左上角坐标
+                - width/height: 显示尺寸
+
+                可选属性:
+                - id: 唯一标识符
+                - rotation: 旋转角度[0, 360)
+                - flipX/flipY: 水平/垂直翻转
+                - alpha: 不透明度[0,1]
+
+                子元素:
+                - svg: 标准 SVG 根元素, 仅描述嵌入内容, 不承载 Slides 布局属性
+                - reflection: 倒影样式, 无reflection标签代表无倒影, 空reflection标签代表使用默认样式
+                - shadow: 阴影样式, 无shadow标签代表无阴影, 空shadow标签代表使用默认样式
+            </xs:documentation>
+        </xs:annotation>
+        <xs:complexType>
+            <xs:sequence>
+                <xs:any namespace="http://www.w3.org/2000/svg" processContents="lax" minOccurs="1" maxOccurs="1"/>
+                <xs:element name="reflection" type="sml:ReflectionType" minOccurs="0"/>
+                <xs:element name="shadow" type="sml:ShadowType" minOccurs="0"/>
+            </xs:sequence>
+
+            <xs:attribute name="id" type="xs:string" use="optional"/>
+            <xs:attribute name="topLeftX" type="sml:XType" use="required"/>
+            <xs:attribute name="topLeftY" type="sml:YType" use="required"/>
+            <xs:attribute name="width" type="sml:PositiveSize" use="required"/>
+            <xs:attribute name="height" type="sml:PositiveSize" use="required"/>
+            <xs:attribute name="rotation" type="sml:RotationType" use="optional" default="0"/>
+            <xs:attribute name="flipX" type="xs:boolean" use="optional" default="false"/>
+            <xs:attribute name="flipY" type="xs:boolean" use="optional" default="false"/>
+            <xs:attribute name="alpha" type="sml:AlphaType" use="optional" default="1"/>
+        </xs:complexType>
+    </xs:element>
+
    <!-- 表格元素 -->
    <xs:element name="table">
        <xs:annotation>
@@ -1500,7 +1708,7 @@
                td 子元素:
                - borderTop/borderRight/borderBottom/borderLeft: 单元格边框样式, 无border标签代表无边框, 空border标签代表使用默认样式(实线边框, 颜色为rgba(221, 222, 223, 1), 宽度为1)
                - fill: 单元格填充样式, 无fill标签代表不填充, 空fill标签代表使用默认样式(默认颜色填充, 颜色为rgba(255, 255, 255, 1))
-                - content: 单元格内容
+                - content: 单元格内容。内容默认不反向修改表格几何尺寸; 当内容高度大于当前行高时, 需要手动修改行高
            </xs:documentation>
        </xs:annotation>
        <xs:complexType>
@@ -1558,13 +1766,15 @@
        <xs:complexType/>
    </xs:element>

-    <xs:element name="strong">
+    <xs:element name="field">
        <xs:annotation>
-            <xs:documentation>粗体/加重文本</xs:documentation>
+            <xs:documentation>
+                动态文本字段。
+                type 属性描述动态语义，元素内容是静态降级文本，可包含行内样式元素。
+            </xs:documentation>
        </xs:annotation>
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
-                <xs:element ref="sml:br"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1574,6 +1784,70 @@
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
            </xs:choice>
+            <xs:attribute name="type" type="sml:FieldType" use="required"/>
+        </xs:complexType>
+    </xs:element>
+
+
+
+    <xs:element name="formula">
+        <xs:annotation>
+            <xs:documentation>
+                通用公式元素。
+                用于展示各类科学公式。
+
+                结构说明：
+                - 必须从支持的公式格式中选择且仅选择一种作为子元素。
+                - 当前版本支持格式：&lt;latex&gt;
+
+                示例：
+                - 基础公式：
+                &lt;formula&gt;
+                &lt;latex&gt;&lt;![CDATA[ E = mc^2 ]]&gt;&lt;/latex&gt;
+                &lt;/formula&gt;
+            </xs:documentation>
+        </xs:annotation>
+        <xs:complexType>
+            <xs:choice minOccurs="1" maxOccurs="1">
+                <xs:element name="latex">
+                    <xs:annotation>
+                        <xs:documentation>
+                            LaTeX 格式的公式内容。
+                            本元素包含的 LaTeX 字符串必须严格符合附件中定义的宏集范围。
+
+                            内容语法：
+                            - 语法范围：仅使用附件白名单中明确支持的宏。
+                            - 表达建议：优先使用基础运算符、分式(\frac)、根号(\sqrt)、矩阵(matrix)等标准数学环境。
+                            - 格式要求：必须使用 CDATA 包裹内容，且 CDATA 内部严禁进行 XML 转义（如 &amp;lt;, &amp;amp;）。
+                            - 空白处理：解析器将保留 CDATA 内的所有换行和缩进，建议利用此特性保持 LaTeX 源码的结构化和可读性。
+                        </xs:documentation>
+                    </xs:annotation>
+                    <xs:simpleType>
+                        <xs:restriction base="xs:string"/>
+                    </xs:simpleType>
+                </xs:element>
+            </xs:choice>
+        </xs:complexType>
+    </xs:element>
+
+    <xs:element name="strong">
+        <xs:annotation>
+            <xs:documentation>粗体/加重文本</xs:documentation>
+        </xs:annotation>
+        <xs:complexType mixed="true">
+            <xs:choice minOccurs="0" maxOccurs="unbounded">
+                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
+                <xs:element ref="sml:strong"/>
+                <xs:element ref="sml:em"/>
+                <xs:element ref="sml:u"/>
+                <xs:element ref="sml:span"/>
+                <xs:element ref="sml:del"/>
+                <xs:element ref="sml:a"/>
+                <xs:element ref="sml:shadow"/>
+                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
+            </xs:choice>
        </xs:complexType>
    </xs:element>

@@ -1590,6 +1864,7 @@
                <xs:extension base="sml:ShadowType">
                    <xs:choice minOccurs="0" maxOccurs="unbounded">
                        <xs:element ref="sml:br"/>
+                        <xs:element ref="sml:formula"/>
                        <xs:element ref="sml:strong"/>
                        <xs:element ref="sml:em"/>
                        <xs:element ref="sml:u"/>
@@ -1598,6 +1873,7 @@
                        <xs:element ref="sml:a"/>
                        <xs:element ref="sml:shadow"/>
                        <xs:element ref="sml:outline"/>
+                        <xs:element ref="sml:field"/>
                    </xs:choice>
                </xs:extension>
            </xs:complexContent>
@@ -1617,6 +1893,7 @@
                <xs:extension base="sml:OutlineType">
                    <xs:choice minOccurs="0" maxOccurs="unbounded">
                        <xs:element ref="sml:br"/>
+                        <xs:element ref="sml:formula"/>
                        <xs:element ref="sml:strong"/>
                        <xs:element ref="sml:em"/>
                        <xs:element ref="sml:u"/>
@@ -1625,6 +1902,7 @@
                        <xs:element ref="sml:a"/>
                        <xs:element ref="sml:shadow"/>
                        <xs:element ref="sml:outline"/>
+                        <xs:element ref="sml:field"/>
                    </xs:choice>
                </xs:extension>
            </xs:complexContent>
@@ -1638,6 +1916,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1646,6 +1925,7 @@
                <xs:element ref="sml:a"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
        </xs:complexType>
    </xs:element>
@@ -1657,6 +1937,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1665,6 +1946,7 @@
                <xs:element ref="sml:a"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
        </xs:complexType>
    </xs:element>
@@ -1676,6 +1958,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1684,6 +1967,7 @@
                <xs:element ref="sml:a"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
        </xs:complexType>
    </xs:element>
@@ -1698,6 +1982,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1706,6 +1991,7 @@
                <xs:element ref="sml:a"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
            <xs:attribute name="color" type="sml:Color" use="optional"/>
            <xs:attribute name="backgroundColor" type="sml:Color" use="optional"/>
@@ -1729,6 +2015,7 @@
        <xs:complexType mixed="true">
            <xs:choice minOccurs="0" maxOccurs="unbounded">
                <xs:element ref="sml:br"/>
+                <xs:element ref="sml:formula"/>
                <xs:element ref="sml:strong"/>
                <xs:element ref="sml:em"/>
                <xs:element ref="sml:u"/>
@@ -1736,6 +2023,7 @@
                <xs:element ref="sml:del"/>
                <xs:element ref="sml:shadow"/>
                <xs:element ref="sml:outline"/>
+                <xs:element ref="sml:field"/>
            </xs:choice>
            <xs:attribute name="href" use="required">
                <xs:simpleType>
@@ -1823,36 +2111,116 @@
    <!-- 有序列表样式枚举 -->
    <xs:simpleType name="OrderedListStyle">
        <xs:annotation>
-            <xs:documentation>有序列表样式</xs:documentation>
+            <xs:documentation>
+                有序列表样式
+
+                分为两类：
+                1. 复合样式（按层级循环不同格式）：如 number-lower-alpha-lower-roman 表示第1级用数字、第2级用小写字母、第3级用小写罗马，超过层级数后循环
+                2. 单一样式（所有层级使用同一格式，不循环）：以 PPTX 标准 scheme 命名，如 alpha-lc-paren-both 表示所有层级都用 (a)(b)(c) 格式
+            </xs:documentation>
        </xs:annotation>
        <xs:restriction base="xs:string">
+            <!-- 复合样式（按层级循环） -->
            <xs:enumeration value="number-lower-alpha-lower-roman"><xs:annotation><xs:documentation>1. a. i. - 数字/小写字母/小写罗马</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="number-lower-alpha-lower-roman-paren"><xs:annotation><xs:documentation>1) a) i) - 带括号版本</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="hierarchical-number"><xs:annotation><xs:documentation>1. 1.1. 1.1.1. - 多级数字</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="upper-alpha-lower-alpha-lower-roman"><xs:annotation><xs:documentation>A. a. i. - 大写字母/小写字母/小写罗马</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="upper-roman-upper-alpha-number"><xs:annotation><xs:documentation>I. A. 1. - 大写罗马/大写字母/数字</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="zero-padded-lower-alpha-lower-roman"><xs:annotation><xs:documentation>01. a. i. - 补零数字/小写字母/小写罗马</xs:documentation></xs:annotation></xs:enumeration>
-            <xs:enumeration value="circle-number"><xs:annotation><xs:documentation> 圆圈数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="circle-number"><xs:annotation><xs:documentation>圆圈数字</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="lower-alpha-paren"><xs:annotation><xs:documentation>a) b) c) - 小写字母带括号</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="lower-alpha-dot"><xs:annotation><xs:documentation>a. b. c. - 小写字母带点</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="chinese-formal"><xs:annotation><xs:documentation>一、二、三、 - 中文数字</xs:documentation></xs:annotation></xs:enumeration>
+
+            <!-- 单一样式（所有层级使用同一格式，不随层级循环） -->
+            <!-- 拉丁字母 Latin -->
+            <xs:enumeration value="alpha-lc-paren-both"><xs:annotation><xs:documentation>(a) (b) (c) - 小写字母带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="alpha-uc-paren-both"><xs:annotation><xs:documentation>(A) (B) (C) - 大写字母带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="alpha-lc-paren-r"><xs:annotation><xs:documentation>a) b) c) - 小写字母带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="alpha-uc-paren-r"><xs:annotation><xs:documentation>A) B) C) - 大写字母带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="alpha-lc-period"><xs:annotation><xs:documentation>a. b. c. - 小写字母带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="alpha-uc-period"><xs:annotation><xs:documentation>A. B. C. - 大写字母带点</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 阿拉伯数字 Arabic Numeral -->
+            <xs:enumeration value="arabic-paren-both"><xs:annotation><xs:documentation>(1) (2) (3) - 数字带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic-paren-r"><xs:annotation><xs:documentation>1) 2) 3) - 数字带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic-period"><xs:annotation><xs:documentation>1. 2. 3. - 数字带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic-plain"><xs:annotation><xs:documentation>1 2 3 - 纯数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic-db-period"><xs:annotation><xs:documentation>１．２．３．- 全角数字带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic-db-plain"><xs:annotation><xs:documentation>１ ２ ３ - 全角纯数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic1-minus"><xs:annotation><xs:documentation>أ- ب- ت- - 阿拉伯语字母(现代序)带后横线</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arabic2-minus"><xs:annotation><xs:documentation>-أ- -ب- -ج- - 阿拉伯语字母(Abjadi序)带双横线</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 罗马数字 Roman -->
+            <xs:enumeration value="roman-lc-paren-both"><xs:annotation><xs:documentation>(i) (ii) (iii) - 小写罗马带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="roman-uc-paren-both"><xs:annotation><xs:documentation>(I) (II) (III) - 大写罗马带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="roman-lc-paren-r"><xs:annotation><xs:documentation>i) ii) iii) - 小写罗马带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="roman-uc-paren-r"><xs:annotation><xs:documentation>I) II) III) - 大写罗马带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="roman-lc-period"><xs:annotation><xs:documentation>i. ii. iii. - 小写罗马带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="roman-uc-period"><xs:annotation><xs:documentation>I. II. III. - 大写罗马带点</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 圆圈数字 Circle -->
+            <xs:enumeration value="circle-num-db-plain"><xs:annotation><xs:documentation>① ② ③ - 圆圈数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="circle-num-wd-black-plain"><xs:annotation><xs:documentation>❶ ❷ ❸ - 实心圆圈数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="circle-num-wd-white-plain"><xs:annotation><xs:documentation>① ② ③ - 圆圈数字（1-10 循环, 字形与 circle-num-db-plain 相同但超过 10 后不降级为纯数字）</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 东亚 East Asian -->
+            <xs:enumeration value="ea1-chs-period"><xs:annotation><xs:documentation>一. 二. 三. - 简体中文带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-chs-plain"><xs:annotation><xs:documentation>一 二 三 - 简体中文</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-cht-period"><xs:annotation><xs:documentation>一. 二. 三. - 繁体中文带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-cht-plain"><xs:annotation><xs:documentation>一 二 三 - 繁体中文</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-jpn-chs-db-period"><xs:annotation><xs:documentation>一．二．三．- CJK汉字数字带全角点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-jpn-kor-plain"><xs:annotation><xs:documentation>一 二 三 - CJK汉字数字</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="ea1-jpn-kor-period"><xs:annotation><xs:documentation>一. 二. 三. - CJK汉字数字带半角点</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 希伯来语 Hebrew -->
+            <xs:enumeration value="hebrew2-minus"><xs:annotation><xs:documentation>א- ב- ג- - 希伯来字母带横线</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 泰语 Thai -->
+            <xs:enumeration value="thai-alpha-period"><xs:annotation><xs:documentation>ก. ข. ค. - 泰语字母带点（跳过 ฃ/ฅ/ฆ）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="thai-alpha-paren-r"><xs:annotation><xs:documentation>ก) ข) ค) - 泰语字母带右括号（跳过 ฃ/ฅ/ฆ）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="thai-alpha-paren-both"><xs:annotation><xs:documentation>(ก) (ข) (ค) - 泰语字母带双括号（跳过 ฃ/ฅ/ฆ）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="thai-num-period"><xs:annotation><xs:documentation>๑. ๒. ๓. - 泰语数字带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="thai-num-paren-r"><xs:annotation><xs:documentation>๑) ๒) ๓) - 泰语数字带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="thai-num-paren-both"><xs:annotation><xs:documentation>(๑) (๒) (๓) - 泰语数字带双括号</xs:documentation></xs:annotation></xs:enumeration>
+            <!-- 印地语 Hindi -->
+            <xs:enumeration value="hindi-alpha-period"><xs:annotation><xs:documentation>अ. आ. इ. - 印地语元音字母带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="hindi-num-period"><xs:annotation><xs:documentation>१. २. ३. - 印地语数字带点</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="hindi-num-paren-r"><xs:annotation><xs:documentation>१) २) ३) - 印地语数字带右括号</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="hindi-alpha1-period"><xs:annotation><xs:documentation>क. ख. ग. - 印地语辅音字母带点</xs:documentation></xs:annotation></xs:enumeration>
        </xs:restriction>
    </xs:simpleType>
    <!-- 无序列表样式枚举 -->
    <xs:simpleType name="UnorderedListStyle">
        <xs:annotation>
-            <xs:documentation>无序列表样式</xs:documentation>
+            <xs:documentation>
+                无序列表样式
+
+                分为两类：
+                1. 复合样式（按层级循环不同图标）：如 circle-hollow-square 表示第1级实心圆、第2级空心圆、第3级实心方形，超过层级数后循环
+                2. 单一样式（所有层级使用同一图标，不循环）：以 pptx- 前缀命名，如 pptx-circle 表示所有层级都用 ● 实心圆
+            </xs:documentation>
        </xs:annotation>
        <xs:restriction base="xs:string">
+            <!-- 复合样式（按层级循环） -->
            <xs:enumeration value="circle-hollow-square"><xs:annotation><xs:documentation>实心圆 空心圆 实心方形</xs:documentation></xs:annotation></xs:enumeration>
-            <xs:enumeration value="diamond-triangle-square"><xs:annotation><xs:documentation>棱形 三角形 实心方形</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="diamond-triangle-square"><xs:annotation><xs:documentation>菱形 三角形 实心方形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="hollow-square-all"><xs:annotation><xs:documentation>空心方形 空心方形 空心方形</xs:documentation></xs:annotation></xs:enumeration>
-            <xs:enumeration value="arrow-diamond-circle"><xs:annotation><xs:documentation>右箭头 实心棱形 实心圆形</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="arrow-diamond-circle"><xs:annotation><xs:documentation>右箭头 实心菱形 实心圆形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="star-hollow-circle-square"><xs:annotation><xs:documentation>实心五角星 空心圆形 实心方形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="triangle-hollow-circle-square"><xs:annotation><xs:documentation>三角形 空心圆形 实心方形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="solid-square-all"><xs:annotation><xs:documentation>实心方形 实心方形 实心方形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="solid-diamond-all"><xs:annotation><xs:documentation>实心菱形 实心菱形 实心菱形</xs:documentation></xs:annotation></xs:enumeration>
            <xs:enumeration value="check-all"><xs:annotation><xs:documentation>对勾 对勾 对勾</xs:documentation></xs:annotation></xs:enumeration>
+
+            <!-- 单一样式（所有层级使用同一图标，不随层级循环） -->
+            <xs:enumeration value="pptx-circle"><xs:annotation><xs:documentation>● 实心圆（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-square"><xs:annotation><xs:documentation>■ 方块（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-diamond"><xs:annotation><xs:documentation>◆ 菱形（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-square-empty"><xs:annotation><xs:documentation>□ 空心方框（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-check"><xs:annotation><xs:documentation>✓ 对勾（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-triangle"><xs:annotation><xs:documentation>► 右三角（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-bullet"><xs:annotation><xs:documentation>• 小圆点（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+
+            <xs:enumeration value="pptx-circle-empty"><xs:annotation><xs:documentation>○ 空心圆（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-diamond-empty"><xs:annotation><xs:documentation>◇ 空心菱形（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-arrow-right"><xs:annotation><xs:documentation>➔ 右箭头（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-star"><xs:annotation><xs:documentation>★ 星形（所有层级）</xs:documentation></xs:annotation></xs:enumeration>
+            <xs:enumeration value="pptx-square-shadow"><xs:annotation><xs:documentation>❑ 带右下阴影的 3D 方框（所有层级，对应 PPTX Wingdings 'q'）</xs:documentation></xs:annotation></xs:enumeration>
        </xs:restriction>
    </xs:simpleType>

@@ -2096,6 +2464,19 @@
        </xs:restriction>
    </xs:simpleType>

+    <xs:simpleType name="ChartGradientKindType">
+        <xs:annotation>
+            <xs:documentation>
+                图表渐变类型
+                可选值: linear(线性渐变) | radial(径向渐变)
+            </xs:documentation>
+        </xs:annotation>
+        <xs:restriction base="xs:string">
+            <xs:enumeration value="linear"/>
+            <xs:enumeration value="radial"/>
+        </xs:restriction>
+    </xs:simpleType>
+
    <xs:simpleType name="ChartRadarShapeType">
        <xs:annotation>
            <xs:documentation>
@@ -2217,6 +2598,7 @@

                属性:
                - textAlign: 文本对齐方式(left|center|right), 默认left
+                - fontFamily: 字体族名称，仅图表根级主标题/副标题支持；坐标轴标题不支持
                - fontSize: 字号大小
                - bold: 是否加粗
                - italic: 是否斜体, 默认false
@@ -2230,6 +2612,7 @@
        <xs:complexContent>
            <xs:extension base="sml:ChartFontStyleType">
                <xs:attribute name="textAlign" type="sml:ChartTextAlignType" use="optional" default="left"/>
+                <xs:attribute name="fontFamily" type="sml:FontFamilyType" use="optional"/>
            </xs:extension>
        </xs:complexContent>
    </xs:complexType>
@@ -2298,10 +2681,10 @@
                图表背景配置

                属性:
-                - color: 背景颜色, 默认透明 rgba(0,0,0,0)
+                - color: 背景颜色，省略时使用图表默认背景；无填充可使用透明色 rgba(0,0,0,0)
            </xs:documentation>
        </xs:annotation>
-        <xs:attribute name="color" type="sml:SolidColor" use="optional" default="rgb(255, 255, 255)"/>
+        <xs:attribute name="color" type="sml:SolidColor" use="optional"/>
    </xs:complexType>

    <xs:complexType name="ChartBorderType">
@@ -2311,7 +2694,7 @@

                属性:
                - color: 边框颜色，默认 rgb(222, 224, 227)
-                - width: 边框宽度(像素), 默认 1
+                - width: 边框宽度(像素), 默认 1；无边框可设置为0，或不设置chartBorder
                - style: 边框样式(solid|dashed|dotted), 默认 solid
                - radius: 圆角半径(像素), 默认 6
            </xs:documentation>
@@ -2322,6 +2705,61 @@
        <xs:attribute name="radius" type="xs:nonNegativeInteger" use="optional" />
    </xs:complexType>

+    <xs:complexType name="ChartGradientStopType">
+        <xs:annotation>
+            <xs:documentation>
+                图表渐变色标
+
+                属性:
+                - offset: 色标位置比例[0,1]
+                - color: 色标颜色
+                - opacity: 色标透明度[0,1]
+            </xs:documentation>
+        </xs:annotation>
+        <xs:attribute name="offset" type="sml:RatioType" use="required"/>
+        <xs:attribute name="color" type="sml:SolidColor" use="required"/>
+        <xs:attribute name="opacity" type="sml:RatioType" use="optional"/>
+    </xs:complexType>
+
+    <xs:complexType name="ChartGradientStopsType">
+        <xs:annotation>
+            <xs:documentation>
+                图表渐变色标列表，至少需要2个色标
+            </xs:documentation>
+        </xs:annotation>
+        <xs:sequence>
+            <xs:element name="stop" type="sml:ChartGradientStopType" minOccurs="2" maxOccurs="unbounded"/>
+        </xs:sequence>
+    </xs:complexType>
+
+    <xs:complexType name="ChartGradientType">
+        <xs:annotation>
+            <xs:documentation>
+                图表渐变配置
+
+                属性:
+                - type: 渐变类型(linear|radial)
+                - x0/y0/x1/y1: 线性渐变起止点坐标
+                - r0/r1: 径向渐变半径
+                - gradientMethod: 渐变算法/插值方式
+
+                子元素:
+                - stops: 渐变色标列表
+            </xs:documentation>
+        </xs:annotation>
+        <xs:sequence>
+            <xs:element name="stops" type="sml:ChartGradientStopsType" minOccurs="1"/>
+        </xs:sequence>
+        <xs:attribute name="type" type="sml:ChartGradientKindType" use="required"/>
+        <xs:attribute name="x0" type="xs:double" use="optional"/>
+        <xs:attribute name="y0" type="xs:double" use="optional"/>
+        <xs:attribute name="x1" type="xs:double" use="optional"/>
+        <xs:attribute name="y1" type="xs:double" use="optional"/>
+        <xs:attribute name="r0" type="xs:double" use="optional"/>
+        <xs:attribute name="r1" type="xs:double" use="optional"/>
+        <xs:attribute name="gradientMethod" type="xs:string" use="optional"/>
+    </xs:complexType>
+
    <xs:complexType name="ChartColorThemeType">
        <xs:annotation>
            <xs:documentation>
@@ -2431,12 +2869,16 @@
                - size: 该系列所有点的大小

                子元素:
+                - fillGradient: 该系列所有点的填充渐变(可选)
+                - strokeGradient: 该系列所有点的边框/描边渐变(可选)
                - chartPoint: 单个数据点配置(可选, 多个), 用于覆盖特定点的样式
            </xs:documentation>
        </xs:annotation>
        <xs:complexContent>
            <xs:extension base="sml:ChartGlobalPointsType">
                <xs:sequence>
+                    <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                    <xs:element name="strokeGradient" type="sml:ChartGradientType" minOccurs="0"/>
                    <xs:element name="chartPoint" minOccurs="0" maxOccurs="unbounded">
                        <xs:complexType>
                            <xs:annotation>
@@ -2448,8 +2890,14 @@
                                    - color: 该点的颜色
                                    - shape: 该点的形状(circle|square|triangle|diamond|rect)
                                    - size: 该点的大小(像素)
+
+                                    子元素:
+                                    - fillGradient: 该点填充渐变(可选)
                                </xs:documentation>
                            </xs:annotation>
+                            <xs:sequence>
+                                <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                            </xs:sequence>
                            <xs:attribute name="index" type="xs:positiveInteger" use="required"/>
                            <xs:attribute name="color" type="sml:SolidColor" use="optional"/>
                            <xs:attribute name="shape" type="sml:ChartPointShapeType" use="optional"/>
@@ -2462,7 +2910,7 @@
    </xs:complexType>

    <!-- 线条配置 -->
-    <xs:complexType name="ChartLineType">
+    <xs:complexType name="ChartGlobalLineType">
        <xs:annotation>
            <xs:documentation>
                图表全局线条配置（第一层：所有系列的默认样式）
@@ -2479,8 +2927,27 @@
        <xs:attribute name="style" type="sml:ChartLineStyleType" use="optional" default="solid"/>
    </xs:complexType>

+    <xs:complexType name="ChartSeriesLineType">
+        <xs:annotation>
+            <xs:documentation>
+                图表系列线条配置（第二层：单系列统一配置）
+                继承ChartGlobalLineType的所有属性
+
+                子元素:
+                - strokeGradient: 该系列线条渐变(可选)
+            </xs:documentation>
+        </xs:annotation>
+        <xs:complexContent>
+            <xs:extension base="sml:ChartGlobalLineType">
+                <xs:sequence>
+                    <xs:element name="strokeGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                </xs:sequence>
+            </xs:extension>
+        </xs:complexContent>
+    </xs:complexType>
+
    <!-- 面积配置 -->
-    <xs:complexType name="ChartAreaType">
+    <xs:complexType name="ChartGlobalAreaType">
        <xs:annotation>
            <xs:documentation>
                图表全局面积配置（第一层：所有系列的默认填充样式）
@@ -2493,6 +2960,25 @@
        <xs:attribute name="color" type="sml:SolidColor" use="optional"/>
    </xs:complexType>

+    <xs:complexType name="ChartSeriesAreaType">
+        <xs:annotation>
+            <xs:documentation>
+                图表系列面积配置（第二层：单系列统一配置）
+                继承ChartGlobalAreaType的所有属性
+
+                子元素:
+                - fillGradient: 该系列面积填充渐变(可选)
+            </xs:documentation>
+        </xs:annotation>
+        <xs:complexContent>
+            <xs:extension base="sml:ChartGlobalAreaType">
+                <xs:sequence>
+                    <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                </xs:sequence>
+            </xs:extension>
+        </xs:complexContent>
+    </xs:complexType>
+
    <!-- 柱子配置 -->
    <xs:complexType name="ChartGlobalBarsType">
        <xs:annotation>
@@ -2533,12 +3019,16 @@
                - borderStyle: 该系列所有柱子的边框样式

                子元素:
+                - fillGradient: 该系列所有柱子的填充渐变(可选)
+                - strokeGradient: 该系列所有柱子的边框渐变(可选)
                - chartBar: 单个柱子配置(可选, 多个), 用于覆盖特定柱子的样式
            </xs:documentation>
        </xs:annotation>
        <xs:complexContent>
            <xs:extension base="sml:ChartGlobalBarsType">
                <xs:sequence>
+                    <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                    <xs:element name="strokeGradient" type="sml:ChartGradientType" minOccurs="0"/>
                    <xs:element name="chartBar" minOccurs="0" maxOccurs="unbounded">
                        <xs:complexType>
                            <xs:annotation>
@@ -2551,8 +3041,14 @@
                                    - borderColor: 该柱子的边框颜色
                                    - borderWidth: 该柱子的边框宽度(像素)
                                    - borderStyle: 该柱子的边框样式(solid|dashed|dotted)
+
+                                    子元素:
+                                    - fillGradient: 该柱子的填充渐变(可选)
                                </xs:documentation>
                            </xs:annotation>
+                            <xs:sequence>
+                                <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+                            </xs:sequence>
                            <xs:attribute name="index" type="xs:positiveInteger" use="required"/>
                            <xs:attribute name="color" type="sml:SolidColor" use="optional"/>
                            <xs:attribute name="borderColor" type="sml:SolidColor" use="optional"/>
@@ -2577,8 +3073,14 @@
                - offsetRadius: 扇区径向偏移比例[0,1], 用于突出显示
                - borderColor: 扇区边框颜色
                - color: 扇区填充颜色
+
+                子元素:
+                - fillGradient: 扇区填充渐变(可选)
            </xs:documentation>
        </xs:annotation>
+        <xs:sequence>
+            <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
+        </xs:sequence>
        <xs:attribute name="index" type="xs:positiveInteger" use="required"/>
        <xs:attribute name="offsetRadius" type="sml:RatioType" use="optional"/>
        <xs:attribute name="borderColor" type="sml:SolidColor" use="optional"/>
@@ -2598,10 +3100,12 @@
                - startAngle: 起始角度[0,360), 控制第一个扇区的起始位置, 默认0

                子元素:
+                - fillGradient: 所有扇区的统一填充渐变(可选)
                - chartSector: 单个扇区配置(可选, 多个), 用于定制特定扇区
            </xs:documentation>
        </xs:annotation>
        <xs:sequence>
+            <xs:element name="fillGradient" type="sml:ChartGradientType" minOccurs="0"/>
            <xs:element name="chartSector" type="sml:ChartSectorType" minOccurs="0" maxOccurs="unbounded"/>
        </xs:sequence>
        <xs:attribute name="borderColor" type="sml:SolidColor" use="optional"/>
@@ -2648,8 +3152,8 @@
        </xs:annotation>
        <xs:sequence>
            <xs:element name="chartPoints" type="sml:ChartSeriesPointsType" minOccurs="0"/>
-            <xs:element name="chartLine" type="sml:ChartLineType" minOccurs="0"/>
-            <xs:element name="chartArea" type="sml:ChartAreaType" minOccurs="0"/>
+            <xs:element name="chartLine" type="sml:ChartSeriesLineType" minOccurs="0"/>
+            <xs:element name="chartArea" type="sml:ChartSeriesAreaType" minOccurs="0"/>
            <xs:element name="chartBars" type="sml:ChartSeriesBarsType" minOccurs="0"/>
            <xs:element name="chartSectors" type="sml:ChartSectorsType" minOccurs="0"/>
            <xs:element name="chartLabels" type="sml:ChartDataLabelsType" minOccurs="0"/>
@@ -2850,8 +3354,8 @@
        </xs:annotation>
        <xs:all>
            <xs:element name="chartPoints" type="sml:ChartGlobalPointsType" minOccurs="0"/>
-            <xs:element name="chartLines" type="sml:ChartLineType" minOccurs="0"/>
-            <xs:element name="chartAreas" type="sml:ChartAreaType" minOccurs="0"/>
+            <xs:element name="chartLines" type="sml:ChartGlobalLineType" minOccurs="0"/>
+            <xs:element name="chartAreas" type="sml:ChartGlobalAreaType" minOccurs="0"/>
            <xs:element name="chartBars" type="sml:ChartGlobalBarsType" minOccurs="0"/>
            <xs:element name="chartLabels" type="sml:ChartDataLabelsType" minOccurs="0"/>
            <xs:element name="chartSeriesList" type="sml:ChartSeriesListType" minOccurs="0"/>
--- a/skills/lark-slides/references/xml-schema-quick-ref.md
+++ b/skills/lark-slides/references/xml-schema-quick-ref.md
@@ -81,7 +81,7 @@ XSD 中的 `title`、`headline`、`sub-headline`、`body`、`caption` 主要出
 **子元素：**

 - `<style>?` - 页面样式，目前可放 `<fill>`
- `<data>?` - 页面元素容器，可放 `shape`、`line`、`polyline`、`img`、`table`、`icon`、`chart`、`undefined`
+- `<data>?` - 页面元素容器，可放 `shape`、`line`、`polyline`、`img`、`table`、`icon`、`embed`、`chart`、`undefined`
 - `<note>?` - 演讲者备注，内部可放 `<content>`

 这意味着 `<title>`、`<headline>`、`<body>`、`<caption>` 不能直接放在 `<slide>` 下。
@@ -129,6 +129,15 @@ XSD 中的 `title`、`headline`、`sub-headline`、`body`、`caption` 主要出
 - `<a>`
 - `<shadow>`
 - `<outline>`
+- `<formula>`
+
+公式写法：
+
+```xml
+<p>公式：<formula><latex><![CDATA[ E = mc^2 ]]></latex></formula></p>
+```
+
+`<formula>` 是内联元素；当前只支持一个 `<latex>` 子元素。LaTeX 内容必须放在 `CDATA` 中，且 `CDATA` 内不要写 XML 转义；宏只使用服务端支持范围内的写法，优先用基础运算符、`\frac`、`\sqrt`、`matrix`。

 示例：

@@ -312,10 +321,54 @@ XSD 中的 `title`、`headline`、`sub-headline`、`body`、`caption` 主要出

 `<chart>` 直接子元素必须有 `<chartPlotArea>`（绘图区）和 `<chartData>`（数据）；`<chartTitle>`、`<chartSubTitle>`、`<chartStyle>`、`<chartLegend>`、`<chartTooltip>` 可选，如果想不展示标题、副标题、图例或悬浮提示，省略相应元素标签即可。

+`<chartStyle>` 常用子元素：
+
+- `<chartBackground>`：`color` 省略时由渲染端决定默认背景；需要完全透明请显式写 `color="rgba(0, 0, 0, 0)"`
+- `<chartBorder>`：无边框可写 `width="0"`，或直接不写 `<chartBorder>` 元素
+
+#### 图表渐变 `<fillGradient>` / `<strokeGradient>`
+
+图表支持渐变填充/描边，`<fillGradient>` 用于面积、柱子、数据点、扇区填充，`<strokeGradient>` 用于线条、数据点边框、柱子边框。渐变只能挂在系列级或单元素级，不要挂在 `<chartPlot>` 全局层。
+
+可挂载位置：
+
+- 系列级：`<chartBars>` / `<chartPoints>` 支持 `<fillGradient>` 与 `<strokeGradient>`；`<chartLine>` 只支持 `<strokeGradient>`；`<chartArea>` / `<chartSectors>` 只支持 `<fillGradient>`
+- 单元素级：`<chartBar index="...">` / `<chartPoint index="...">` / `<chartSector index="...">` 只支持 `<fillGradient>`
+- 全局级：`<chartPlot>` 下的 `<chartLines>` / `<chartAreas>` / `<chartBars>` / `<chartPoints>` 不支持渐变
+
+结构要点：`type` 必填，可为 `linear` 或 `radial`；`linear` 用 `x0` / `y0` / `x1` / `y1`，`radial` 用 `r0` / `r1`；`<stops>` 至少包含 2 个 `<stop>`，`offset` 与 `opacity` 取值均为 `[0, 1]`。
+
+```xml
+<chartSeries index="1">
+  <chartBars>
+    <fillGradient type="linear" x0="0" y0="0" x1="0" y1="1">
+      <stops>
+        <stop offset="0" color="rgb(28, 71, 120)"/>
+        <stop offset="1" color="rgb(28, 71, 120)" opacity="0.3"/>
+      </stops>
+    </fillGradient>
+  </chartBars>
+</chartSeries>
+```
+
 隐藏 `<chart>` 的图例只能通过不写或删除 `<chartLegend>` 实现，`<chartLegend>` 不支持 `position="none"`。

 详细用法见 [slides_xml_schema_definition.xml](slides_xml_schema_definition.xml)。

+### embed
+
+嵌入内容容器：外层 `<embed>` 承载 Slides 的摆放和效果属性（`topLeftX`/`topLeftY`/`width`/`height` 必填，`rotation`/`flipX`/`flipY`/`alpha` 可选），内层承载外部标准内容。当前内层内容为标准 SVG，`<svg>` 必须使用 `http://www.w3.org/2000/svg` 命名空间，且仅描述嵌入内容本身，不承载 Slides 布局属性。
+
+```xml
+<embed topLeftX="80" topLeftY="120" width="200" height="120">
+  <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 200 120">
+    <circle cx="100" cy="60" r="40" fill="rgba(37, 99, 235, 1)"/>
+  </svg>
+</embed>
+```
+
+`<embed>` 直接子元素为一个 `<svg>`（必需），以及可选的 `<reflection>`（倒影）与 `<shadow>`（阴影）。
+
 ## 颜色与样式

 ### fill
--- a/skills/lark-slides/scripts/xml_text_overlap_lint.py
+++ b/skills/lark-slides/scripts/xml_text_overlap_lint.py
@@ -49,6 +49,17 @@ ROUNDTRIP_SXSD_ATTRS = {
 ROUNDTRIP_SXSD_TAGS = {"chartParsedValues"}
 DEFAULT_TABLE_COLUMN_WIDTH = 110
 DEFAULT_TABLE_ROW_HEIGHT = 37
+DEFAULT_TEXT_LINE_SPACING_MULTIPLE = 1.5
+TEXT_WRAP_WIDTH_TOLERANCE_PX = 1.0
+TEXT_HEIGHT_OVERFLOW_TOLERANCE_PX = 0.5
+SINGLE_LINE_METRIC_WIDTH_RATIO = 1.18
+CENTERED_SHORT_LABEL_WIDTH_RATIO = 1.12
+HEADLINE_NEAR_FIT_WIDTH_RATIO = 1.04
+DENSE_BODY_LINE_SPACING_MAX_MULTIPLE = 1.6
+GHOST_TEXT_MIN_FONT_SIZE = 96
+GHOST_TEXT_MAX_ALPHA = 0.5
+GHOST_TEXT_FAINT_MIN_FONT_SIZE = 36
+GHOST_TEXT_FAINT_MAX_ALPHA = 0.35
 # Sub-pixel canvas overflow is floating-point rounding noise (e.g. rotated-bbox math), not a
 # visible defect; keep this well under 1px so real overflow is still always caught.
 CANVAS_OVERFLOW_TOLERANCE = 0.5
@@ -106,6 +117,52 @@ def extract_numeric_attribute(tag_source: str, name: str) -> int | float | None:
    return int(value) if value.is_integer() else value


+def extract_bool_attribute(tag_source: str, name: str) -> bool:
+    value = extract_attribute(tag_source, name)
+    return value in {"true", "1", "yes"}
+
+
+def extract_color_alpha(color: str | None) -> int | float | None:
+    if color is None:
+        return None
+    normalized = re.sub(r"\s+", "", color).lower()
+    if normalized == "transparent":
+        return 0
+    rgba_match = re.fullmatch(
+        r"rgba\([^,]+,[^,]+,[^,]+,([+-]?(?:[0-9]+(?:\.[0-9]*)?|\.[0-9]+))\)",
+        normalized,
+    )
+    if rgba_match is None:
+        return None
+    try:
+        alpha = float(rgba_match.group(1))
+    except ValueError:
+        return None
+    return int(alpha) if alpha.is_integer() else alpha
+
+
+def effective_text_alpha(shape_alpha: int | float | None, text_color: str | None) -> int | float:
+    base_alpha = shape_alpha if isinstance(shape_alpha, (int, float)) else 1
+    color_alpha = extract_color_alpha(text_color)
+    if not isinstance(color_alpha, (int, float)):
+        return base_alpha
+    return base_alpha * color_alpha
+
+
+def detect_inline_style_presence(content_xml: str, style_tags: set[str]) -> bool:
+    for tag_name in style_tags:
+        if re.search(fr"<{re.escape(tag_name)}\b[\s>]", content_xml) is not None:
+            return True
+    return False
+
+
+def detect_any_span_bool_attribute(content_xml: str, attr_name: str) -> bool:
+    for attrs in re.findall(r"<span\b([^>]*)>", content_xml):
+        if extract_bool_attribute(attrs, attr_name):
+            return True
+    return False
+
+
 def sum_sizes(sizes: list[int | float]) -> int | float:
    return sum(sizes)

@@ -218,6 +275,7 @@ def extract_text_paragraphs(value: str, default_font_size: int | float) -> list[
                "lineSpacing": extract_attribute(attrs, "lineSpacing"),
                "beforeLineSpacing": extract_attribute(attrs, "beforeLineSpacing"),
                "afterLineSpacing": extract_attribute(attrs, "afterLineSpacing"),
+                "letterSpacing": extract_numeric_attribute(attrs, "letterSpacing"),
            }
        )
    return paragraphs
@@ -345,7 +403,7 @@ def build_sxsd_tag_hint(tag_name: str, supported_tags: set[str]) -> str:
    if alias:
        return f"Use {alias} instead of <{tag_name}>."
    if tag_name == "svg":
-        return 'Inside <whiteboard>, write SVG as <svg xmlns="http://www.w3.org/2000/svg">...</svg>.'
+        return 'Inside <embed> or <whiteboard>, write SVG as <svg xmlns="http://www.w3.org/2000/svg">...</svg>.'
    close_matches = get_close_matches(tag_name, sorted(supported_tags), n=3, cutoff=0.72)
    if close_matches:
        return "Unsupported SXSD tag. Did you mean " + ", ".join(f"<{match}>" for match in close_matches) + "?"
@@ -366,7 +424,7 @@ def build_sxsd_attr_hint(tag_name: str, attr_name: str, allowed_attrs: set[str])


 def should_skip_sxsd_subtree(element: ET.Element, ancestors: list[str]) -> bool:
-    return "whiteboard" in ancestors and xml_namespace(element.tag) == SVG_NS
+    return ("whiteboard" in ancestors or "embed" in ancestors) and xml_namespace(element.tag) == SVG_NS


 def should_skip_sxsd_attribute(tag_name: str, attr_name: str) -> bool:
@@ -644,7 +702,7 @@ def parse_presentation(xml: str) -> dict[str, Any]:
 def extract_elements(slide_xml: str) -> list[dict[str, Any]]:
    elements: list[dict[str, Any]] = []

-    for match in re.finditer(r"<(shape|img|table|chart|whiteboard)\b([^>]*)>", slide_xml):
+    for match in re.finditer(r"<(shape|img|table|chart|whiteboard|embed)\b([^>]*)>", slide_xml):
        kind, attrs = match.group(1), match.group(2)
        is_self_closing = attrs.rstrip().endswith("/")
        content = ""
@@ -694,6 +752,20 @@ def extract_elements(slide_xml: str) -> list[dict[str, Any]]:
                font_size = extract_numeric_attribute(content_attrs, "fontSize")
                if font_size is None:
                    font_size = extract_numeric_attribute(attrs, "fontSize")
+                font_family = extract_attribute(content_attrs, "fontFamily") or extract_attribute(attrs, "fontFamily")
+                text_color = extract_attribute(content_attrs, "color") or extract_attribute(attrs, "color")
+                bold = (
+                    extract_bool_attribute(content_attrs, "bold")
+                    or extract_bool_attribute(attrs, "bold")
+                    or detect_inline_style_presence(content, {"strong", "b"})
+                    or detect_any_span_bool_attribute(content, "bold")
+                )
+                italic = (
+                    extract_bool_attribute(content_attrs, "italic")
+                    or extract_bool_attribute(attrs, "italic")
+                    or detect_inline_style_presence(content, {"i", "em"})
+                    or detect_any_span_bool_attribute(content, "italic")
+                )
                element.update(
                    {
                        "textType": extract_attribute(content_attrs, "textType"),
@@ -705,11 +777,17 @@ def extract_elements(slide_xml: str) -> list[dict[str, Any]]:
                        "lineSpacing": extract_attribute(content_attrs, "lineSpacing"),
                        "beforeLineSpacing": extract_attribute(content_attrs, "beforeLineSpacing"),
                        "afterLineSpacing": extract_attribute(content_attrs, "afterLineSpacing"),
+                        "letterSpacing": extract_numeric_attribute(content_attrs, "letterSpacing"),
                        "paddingTop": extract_numeric_attribute(content_attrs, "paddingTop") or 0,
                        "paddingRight": extract_numeric_attribute(content_attrs, "paddingRight") or 0,
                        "paddingBottom": extract_numeric_attribute(content_attrs, "paddingBottom") or 0,
                        "paddingLeft": extract_numeric_attribute(content_attrs, "paddingLeft") or 0,
                        "fontSize": font_size if font_size is not None else 16,
+                        "fontFamily": font_family or "",
+                        "color": text_color,
+                        "textAlpha": effective_text_alpha(alpha, text_color),
+                        "bold": bold,
+                        "italic": italic,
                        "text": strip_xml_paragraphs(content),
                        "paragraphs": extract_text_paragraphs(content, font_size if font_size is not None else 16),
                    }
@@ -745,7 +823,11 @@ def is_vertical_text(element: dict[str, Any]) -> bool:

 def detect_image_text_occlusions(elements: list[dict[str, Any]]) -> list[dict[str, Any]]:
    issues: list[dict[str, Any]] = []
-    text_elements = [element for element in elements if is_text_element(element) and has_text_content(element)]
+    text_elements = [
+        element
+        for element in elements
+        if is_text_element(element) and has_text_content(element) and not is_ghost_text(element)
+    ]
    image_elements = [element for element in elements if element["kind"] == "img" and element["alpha"] > 0]
    for text_element in text_elements:
        for image_element in image_elements:
@@ -782,22 +864,135 @@ def normalize_text_for_overlap(text: str) -> str:
    return re.sub(r"\s+", "", text)


-def estimate_character_width(character: str, font_size: int | float) -> int | float:
+SERIF_FONT_PATTERNS = {
+    "song", "songti", "simsun", "ming", "mincho",
+    "georgia", "times", "caslon", "garamond", "sourcehan-serif",
+    "source han serif", "思源宋体", "宋体", "明体",
+}
+
+SANS_EXPLICIT_MARKERS = {"sans", "sans-serif", "sans serif", "sourcehan-sans", "source han sans", "思源黑体", "黑体",
+                         "helvetica", "arial", "inter", "roboto", "verdana", "tahoma", "calibri", "open sans"}
+
+
+def classify_font_family(font_family: str | None) -> str:
+    if not font_family:
+        return "sans"
+    family_lower = font_family.lower()
+    for marker in SANS_EXPLICIT_MARKERS:
+        if marker in family_lower:
+            return "sans"
+    serif_keywords = SERIF_FONT_PATTERNS | {"serif"}
+    for pattern in serif_keywords:
+        if pattern in family_lower:
+            return "serif"
+    return "sans"
+
+
+_FONT_CATEGORY_MULTIPLIERS: dict[str, dict[str, float]] = {
+    "sans": {"upper": 0.57, "lower": 0.51, "digit": 0.58, "punct": 0.50},
+    "serif": {"upper": 0.57, "lower": 0.53, "digit": 0.58, "punct": 0.50},
+}
+
+
+def estimate_character_width(
+    character: str,
+    font_size: int | float,
+    bold: bool = False,
+    font_family: str | None = None,
+) -> int | float:
+    bold_multiplier = 1.05 if bold else 1.0
    if character.isspace():
-        return font_size * 0.33
-    if unicodedata.east_asian_width(character) in {"F", "W"}:
-        return font_size
-    return font_size * 0.55
+        return font_size * 0.33 * bold_multiplier
+    ea_width = unicodedata.east_asian_width(character)
+    if ea_width in {"F", "W"}:
+        return font_size * bold_multiplier
+    category = classify_font_family(font_family)
+    coeffs = _FONT_CATEGORY_MULTIPLIERS[category]
+    if character.isupper():
+        return font_size * coeffs["upper"] * bold_multiplier
+    if character.islower():
+        return font_size * coeffs["lower"] * bold_multiplier
+    if character.isdigit():
+        return font_size * coeffs["digit"] * bold_multiplier
+    return font_size * coeffs["punct"] * bold_multiplier


-def estimate_text_width(text: str, font_size: int | float) -> int | float:
-    return sum(estimate_character_width(character, font_size) for character in text)
+def estimate_text_width(
+    text: str,
+    font_size: int | float,
+    letter_spacing: int | float = 0,
+    bold: bool = False,
+    font_family: str | None = None,
+) -> int | float:
+    base = sum(estimate_character_width(character, font_size, bold, font_family) for character in text)
+    return base + max(len(text) - 1, 0) * letter_spacing
+
+
+def resolve_letter_spacing(element: dict[str, Any], paragraph: dict[str, Any] | None = None) -> int | float:
+    if paragraph is not None:
+        value = paragraph.get("letterSpacing")
+        if isinstance(value, (int, float)):
+            return value
+    value = element.get("letterSpacing")
+    return value if isinstance(value, (int, float)) else 0
+
+
+def text_wrap_width_tolerance() -> int | float:
+    return TEXT_WRAP_WIDTH_TOLERANCE_PX
+
+
+def text_height_overflow_tolerance() -> int | float:
+    return TEXT_HEIGHT_OVERFLOW_TOLERANCE_PX
+
+
+def has_explicit_height_auto_fit(element: dict[str, Any]) -> bool:
+    return element.get("autoFit") in {"normal-auto-fit", "shape-auto-fit"}
+
+
+def is_short_metric_text(text: str) -> bool:
+    compact = re.sub(r"\s+", "", text)
+    if not compact or len(compact) > 16 or re.search(r"\d", compact) is None:
+        return False
+    if re.fullmatch(r"[+\-–—]?[0-9,.，]+[\u4e00-\u9fffA-Za-z]{1,4}", compact):
+        return True
+    if re.search(r"[,.，+\-–—/%％]", compact) is None:
+        return False
+    return re.fullmatch(r"[+\-–—]?[0-9A-Za-z,.，/%％\-–—\u4e00-\u9fff]+", compact) is not None
+
+
+def is_single_line_visual_candidate(
+    element: dict[str, Any],
+    paragraph: dict[str, Any] | None,
+    text: str,
+    logical_width: int | float,
+    effective_width: int | float,
+) -> bool:
+    if "\n" in text or logical_width <= effective_width:
+        return False
+    if is_short_metric_text(text):
+        return logical_width <= effective_width * SINGLE_LINE_METRIC_WIDTH_RATIO
+
+    text_align = (paragraph or {}).get("textAlign") or element.get("textAlign")
+    compact_len = len(re.sub(r"\s+", "", text))
+    if text_align == "center" and compact_len <= 32:
+        return logical_width <= effective_width * CENTERED_SHORT_LABEL_WIDTH_RATIO
+
+    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
+    if element.get("textType") in {"headline", "title"} and font_size <= 30 and compact_len <= 40:
+        return logical_width <= effective_width * HEADLINE_NEAR_FIT_WIDTH_RATIO
+    return False


 def estimate_text_max_line_width(element: dict[str, Any]) -> int | float:
    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
+    bold = element.get("bold", False)
+    font_family = element.get("fontFamily", "")
+    letter_spacing = resolve_letter_spacing(element)
    paragraphs = [paragraph for paragraph in re.split(r"\n+", element["text"]) if paragraph]
-    return max([estimate_text_width(paragraph, font_size) for paragraph in paragraphs] or [1])
+    return max(
+        [estimate_text_width(paragraph, font_size, letter_spacing, bold, font_family) for paragraph in paragraphs]
+        or [1]
+    )


 def is_similar_text_overlay(left: dict[str, Any], right: dict[str, Any]) -> bool:
@@ -810,8 +1005,14 @@ def is_similar_text_overlay(left: dict[str, Any], right: dict[str, Any]) -> bool
    return SequenceMatcher(None, left_text, right_text).ratio() >= 0.75


-def estimate_text_line_count_for_text(element: dict[str, Any], text: str) -> int:
+def estimate_text_line_count_for_text(
+    element: dict[str, Any], text: str, paragraph: dict[str, Any] | None = None
+) -> int:
    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
+    bold = element.get("bold", False)
+    font_family = element.get("fontFamily", "")
+    letter_spacing = resolve_letter_spacing(element, paragraph)
+    available_width = max(element["width"] - element.get("paddingLeft", 0) - element.get("paddingRight", 0), 1)
    hard_lines = text.split("\n")
    if not text:
        return 0
@@ -820,8 +1021,12 @@ def estimate_text_line_count_for_text(element: dict[str, Any], text: str) -> int
        if element.get("wrap") in {"false", "0"}:
            line_count += 1
            continue
-        logical_width = max(estimate_text_width(hard_line, font_size), 1)
-        line_count += max(1, math.ceil(logical_width / max(element["width"], 1)))
+        logical_width = max(estimate_text_width(hard_line, font_size, letter_spacing, bold, font_family), 1)
+        effective_width = available_width + text_wrap_width_tolerance()
+        if is_single_line_visual_candidate(element, paragraph, hard_line, logical_width, effective_width):
+            line_count += 1
+            continue
+        line_count += max(1, math.ceil(logical_width / effective_width))
    return line_count


@@ -831,7 +1036,8 @@ def estimate_text_line_count(element: dict[str, Any]) -> int:

 def estimate_text_line_height(element: dict[str, Any], line_spacing: str | None = None) -> int | float | None:
    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
-    line_spacing = line_spacing or "multiple:1.5"
+    if line_spacing is None:
+        return font_size * DEFAULT_TEXT_LINE_SPACING_MULTIPLE
    match = re.fullmatch(r"(multiple|fixed):([0-9]+(?:\.[0-9]+)?)", line_spacing)
    if match is None:
        return None
@@ -839,12 +1045,27 @@ def estimate_text_line_height(element: dict[str, Any], line_spacing: str | None
    return font_size * float(value) if spacing_type == "multiple" else float(value)


+def adjust_dense_body_line_height(
+    element: dict[str, Any],
+    line_spacing: str | None,
+    line_height: int | float,
+    paragraph_count: int,
+) -> int | float:
+    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
+    if paragraph_count < 4 or font_size > 14 or not line_spacing:
+        return line_height
+    match = re.fullmatch(r"multiple:([0-9]+(?:\.[0-9]+)?)", line_spacing)
+    if match is None:
+        return line_height
+    return min(line_height, font_size * min(float(match.group(1)), DENSE_BODY_LINE_SPACING_MAX_MULTIPLE))
+
+
 def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict[str, Any]]:
    issues: list[dict[str, Any]] = []
    for element in elements:
        if not is_text_element(element) or not has_text_content(element):
            continue
-        if element.get("autoFit") in {"normal-auto-fit", "shape-auto-fit"}:
+        if has_explicit_height_auto_fit(element):
            continue

        font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
@@ -860,10 +1081,11 @@ def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict
        estimated_height = 0.0
        line_heights: list[int | float] = []
        for paragraph in paragraphs:
-            paragraph_line_count = estimate_text_line_count_for_text(element, paragraph["text"])
+            paragraph_line_count = estimate_text_line_count_for_text(element, paragraph["text"], paragraph)
            if paragraph_line_count == 0:
                continue
-            line_height = estimate_text_line_height(element, paragraph["lineSpacing"] or element["lineSpacing"])
+            resolved_line_spacing = paragraph["lineSpacing"] or element["lineSpacing"]
+            line_height = estimate_text_line_height(element, resolved_line_spacing)
            before_spacing = estimate_text_line_height(
                element, paragraph["beforeLineSpacing"] or element["beforeLineSpacing"] or "fixed:0"
            )
@@ -873,6 +1095,7 @@ def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict
            if line_height is None or before_spacing is None or after_spacing is None:
                line_count = 0
                break
+            line_height = adjust_dense_body_line_height(element, resolved_line_spacing, line_height, len(paragraphs))
            first_line_height = font_size if line_count == 0 else line_height
            line_count += paragraph_line_count
            line_heights.append(line_height)
@@ -883,12 +1106,24 @@ def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict
            continue
        available_height = max(element["height"] - element["paddingTop"] - element["paddingBottom"], 0)
        overflow = estimated_height - available_height
-        if overflow <= 0:
+        if overflow <= text_height_overflow_tolerance():
            continue

+        is_background = is_background_decorative_text(element, elements)
+        if is_background:
+            level = "info"
+        else:
+            level = "error" if overflow > 10 else "warning"
+        message = (
+            f'text shape {element["id"]} may overflow its own content box '
+            f'(estimated {estimated_height:g}px, available {available_height:g}px); '
+            'consider setting content wrap="true" autoFit="normal-auto-fit"'
+        )
+        if is_background:
+            message += " (likely background decoration: large font, low alpha, underneath other text)"
        issues.append(
            {
-                "level": "warning",
+                "level": level,
                "code": "text_may_overflow_shape",
                "elements": [element["id"]],
                "line_count": line_count,
@@ -896,11 +1131,7 @@ def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict
                "estimated_height": estimated_height,
                "available_height": available_height,
                "overflow": overflow,
-                "message": (
-                    f'text shape {element["id"]} may overflow its own content box '
-                    f'(estimated {estimated_height:g}px, available {available_height:g}px); '
-                    'consider setting content wrap="true" autoFit="normal-auto-fit"'
-                ),
+                "message": message,
                "hint": (
                    "Increase shape.height, reduce the text, or set content wrap=\"true\" "
                    "autoFit=\"normal-auto-fit\". "
@@ -911,6 +1142,38 @@ def detect_text_may_overflow_shapes(elements: list[dict[str, Any]]) -> list[dict
    return issues


+def is_background_decorative_text(
+    element: dict[str, Any], elements: list[dict[str, Any]]
+) -> bool:
+    if not is_ghost_text(element):
+        return False
+    for other in elements:
+        if other is element:
+            continue
+        if not is_text_element(other) or not has_text_content(other):
+            continue
+        foreground_alpha = other.get("textAlpha", other.get("alpha", 1))
+        if not isinstance(foreground_alpha, (int, float)) or foreground_alpha <= 0:
+            continue
+        if other["order"] <= element["order"]:
+            continue
+        if intersects(element, other):
+            return True
+    return False
+
+
+def is_ghost_text(element: dict[str, Any]) -> bool:
+    if not is_text_element(element) or not has_text_content(element):
+        return False
+    font_size = element["fontSize"] if isinstance(element["fontSize"], (int, float)) else 16
+    text_alpha = element.get("textAlpha", element.get("alpha", 1))
+    if not isinstance(text_alpha, (int, float)):
+        return False
+    if font_size > GHOST_TEXT_MIN_FONT_SIZE and text_alpha < GHOST_TEXT_MAX_ALPHA:
+        return True
+    return font_size >= GHOST_TEXT_FAINT_MIN_FONT_SIZE and text_alpha < GHOST_TEXT_FAINT_MAX_ALPHA
+
+
 def estimate_text_visual_bbox(element: dict[str, Any]) -> dict[str, int | float] | None:
    if not is_text_element(element) or not has_text_content(element) or is_decorative_text(element):
        return None
@@ -1022,6 +1285,8 @@ def should_flag_horizontal_text_overflow(left: dict[str, Any], right: dict[str,
        return False
    if not (has_text_content(left) and has_text_content(right)):
        return False
+    if is_ghost_text(left) or is_ghost_text(right):
+        return False
    if is_template_text_stack(left, right) or is_similar_text_overlay(left, right):
        return False

@@ -1038,13 +1303,16 @@ def should_flag_horizontal_text_overflow(left: dict[str, Any], right: dict[str,
        return False

    font_size = source["fontSize"] if isinstance(source["fontSize"], (int, float)) else 16
+    padding_left = source.get("paddingLeft", 0)
+    padding_right = source.get("paddingRight", 0)
+    available_width = max(source["width"] - padding_left - padding_right, 1)
    visual_width = estimate_text_max_line_width(source)
-    overflow_width = visual_width - source["width"]
-    min_overflow = max(font_size * 1.5, source["width"] * 0.08)
+    overflow_width = visual_width - available_width
+    min_overflow = max(font_size * 1.5, available_width * 0.08)
    if overflow_width < min_overflow:
        return False

-    intrusion_width = source["x"] + visual_width - target["x"]
+    intrusion_width = source["x"] + padding_left + visual_width - target["x"]
    min_intrusion = max(font_size * 1.5, target["width"] * 0.08)
    if intrusion_width < min_intrusion:
        return False
@@ -1056,8 +1324,9 @@ def should_flag_horizontal_text_overflow(left: dict[str, Any], right: dict[str,

 def horizontal_text_overflow_measurement(left: dict[str, Any], right: dict[str, Any]) -> dict[str, int | float]:
    source, target = sorted([left, right], key=lambda element: element["x"])
+    padding_left = source.get("paddingLeft", 0)
    visual_width = estimate_text_max_line_width(source)
-    source_visual_bbox = {"x": source["x"], "y": source["y"], "width": visual_width, "height": source["height"]}
+    source_visual_bbox = {"x": source["x"] + padding_left, "y": source["y"], "width": visual_width, "height": source["height"]}
    width = intersection_width(source_visual_bbox, target)
    height = intersection_height(source_visual_bbox, target)
    return {
@@ -1072,6 +1341,8 @@ def should_flag_overlap(left: dict[str, Any], right: dict[str, Any]) -> bool:
        return False
    if is_text_element(right) and not has_text_content(right):
        return False
+    if is_ghost_text(left) or is_ghost_text(right):
+        return False
    if is_template_text_stack(left, right):
        return False
    if is_text_element(left) and is_text_element(right):
@@ -1118,6 +1389,8 @@ def should_report_whiteboard_overlap(
 ) -> dict[str, Any] | None:
    if other is whiteboard or not intersects(whiteboard, other):
        return None
+    if is_ghost_text(other):
+        return None
    if contains(whiteboard, other):
        return None
    if is_bottom_layer_full_slide_whiteboard(whiteboard, other, slide_width, slide_height):
@@ -1194,6 +1467,8 @@ def detect_whiteboard_external_overlaps(

 def element_canvas_bbox(element: dict[str, Any]) -> dict[str, int | float]:
    bbox = {key: element[key] for key in ("x", "y", "width", "height")}
+    if element["kind"] != "chart" and not (element["kind"] == "shape" and element["type"] == "text"):
+        return bbox
    rotation = element["rotation"]
    if not isinstance(rotation, (int, float)) or not math.isfinite(rotation):
        rotation = 0
@@ -1219,7 +1494,14 @@ def detect_elements_out_of_canvas(
    elements: list[dict[str, Any]], slide_width: int | float, slide_height: int | float
 ) -> list[dict[str, Any]]:
    issues: list[dict[str, Any]] = []
-    for element in elements:
+    for element in (
+        element
+        for element in elements
+        if element["kind"] in {"table", "chart"}
+        or (element["kind"] == "shape" and element["type"] in {"rect", "text"})
+    ):
+        if is_ghost_text(element):
+            continue
        bbox = element_canvas_bbox(element)
        overflow = {
            "left": max(-bbox["x"], 0),
@@ -1635,7 +1917,7 @@ def slide_content_visual_bbox(
        # a straight horizontal/vertical line has zero width or height in one axis; clipped_bbox
        # treats zero-area rects as invisible, so pad to its rendered stroke thickness instead.
        return clipped_bbox(line_stroke_bbox(element), slide_bbox)
-    if element["kind"] in {"img", "chart", "table", "whiteboard", "icon", "polyline"}:
+    if element["kind"] in {"img", "chart", "table", "whiteboard", "embed", "icon", "polyline"}:
        return clipped_bbox(element, slide_bbox)
    return None

@@ -1673,7 +1955,7 @@ def is_slide_content_present(


 def is_large_visual_child(element: dict[str, Any], container: dict[str, Any]) -> bool:
-    if element["kind"] not in {"img", "chart", "table", "whiteboard"}:
+    if element["kind"] not in {"img", "chart", "table", "whiteboard", "embed"}:
        return False
    if not is_visually_rendered(element):
        return False
@@ -1976,8 +2258,6 @@ def normalize_issue(
    elements_by_id: dict[str, dict[str, Any]],
 ) -> dict[str, Any]:
    normalized = dict(issue)
-    if normalized.get("level") == "info":
-        normalized["level"] = "warning"
    element_ids = list(dict.fromkeys(normalized.get("elements", [])))
    normalized["schema_version"] = "2.0"
    normalized["element_ids"] = element_ids
@@ -2039,8 +2319,10 @@ def build_result(
 ) -> dict[str, Any]:
    document_errors = [issue for issue in top_level_issues if issue["level"] == "error"]
    document_warnings = [issue for issue in top_level_issues if issue["level"] == "warning"]
+    document_infos = [issue for issue in top_level_issues if issue["level"] == "info"]
    error_count = len(document_errors) + sum(len(slide["errors"]) for slide in slides)
    warning_count = len(document_warnings) + sum(len(slide["warnings"]) for slide in slides)
+    info_count = len(document_infos) + sum(len(slide["infos"]) for slide in slides)
    all_errors = document_errors + [issue for slide in slides for issue in slide["errors"]]
    all_warnings = document_warnings + [issue for slide in slides for issue in slide["warnings"]]
    status = slide_status(all_errors, all_warnings)
@@ -2053,6 +2335,7 @@ def build_result(
            "slide_count": len(slides),
            "error_count": error_count,
            "warning_count": warning_count,
+            "info_count": info_count,
            "status": status,
            "release_ready": error_count == 0,
            "screenshot_review_required": warning_count > 0,
@@ -2060,6 +2343,7 @@ def build_result(
        "document": {
            "errors": document_errors,
            "warnings": document_warnings,
+            "infos": document_infos,
        },
        "slides": slides,
    }
@@ -2150,6 +2434,7 @@ def lint_xml(xml: str, source_path: str | None = None) -> dict[str, Any]:
        ]
        errors = [issue for issue in issues if issue["level"] == "error"]
        warnings = [issue for issue in issues if issue["level"] == "warning"]
+        infos = [issue for issue in issues if issue["level"] == "info"]
        slides.append(
            {
                "slide_number": slide_number,
@@ -2157,6 +2442,7 @@ def lint_xml(xml: str, source_path: str | None = None) -> dict[str, Any]:
                "element_count": len(elements_by_id),
                "errors": errors,
                "warnings": warnings,
+                "infos": infos,
                "issues": issues,
            }
        )
--- a/skills/lark-slides/scripts/xml_text_overlap_lint_test.py
+++ b/skills/lark-slides/scripts/xml_text_overlap_lint_test.py
@@ -265,6 +265,27 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        self.assertEqual(result["summary"]["error_count"], 0)
        self.assertEqual(result["summary"]["warning_count"], 0)

+    def test_lint_xml_allows_svg_subtree_inside_embed(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <embed topLeftX="80" topLeftY="120" width="240" height="140">
+                  <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 240 140">
+                    <rect x="10" y="10" width="220" height="120" rx="12" fill="#EFF6FF"/>
+                    <circle cx="70" cy="70" r="34" fill="#2563EB"/>
+                    <text x="130" y="76" font-size="18" fill="#1E3A8A">SVG OK</text>
+                  </svg>
+                </embed>
+              </data>
+            </slide>
+            """
+        )
+        codes = [issue["code"] for issue in result.get("issues", [])]
+        self.assertNotIn("sxsd_unsupported_tag", codes)
+        self.assertNotIn("sxsd_unsupported_attr", codes)
+        self.assertEqual(result["summary"]["error_count"], 0)
+
    def test_lint_xml_reports_sxsd_unsupported_tag_with_alias_hint(self) -> None:
        cases = [
            ("textbox", '<textbox topLeftX="80" topLeftY="80" width="300" height="60">Text</textbox>', '<shape type="text">'),
@@ -343,6 +364,26 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        self.assertEqual(result["summary"]["error_count"], 0)
        self.assertNotIn("issues", result)

+    def test_lint_xml_ignores_chart_parsed_values_roundtrip_tag(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <chart topLeftX="80" topLeftY="80" width="300" height="160">
+                  <chartData>
+                    <chartField>
+                      <chartParsedValues>Africa</chartParsedValues>
+                    </chartField>
+                  </chartData>
+                </chart>
+              </data>
+            </slide>
+            """
+        )
+
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertNotIn("issues", result)
+
    def test_lint_xml_limits_chart_roundtrip_attrs_to_matching_tags(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
@@ -637,9 +678,10 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </slide>
            """
        )
-        self.assertEqual(result["summary"]["error_count"], 0)
-        self.assertEqual(result["summary"]["warning_count"], 1)
+        self.assertEqual(result["summary"]["error_count"], 1)
+        self.assertEqual(result["summary"]["warning_count"], 0)
        self.assertEqual(result["slides"][0]["issues"][0]["code"], "text_may_overflow_shape")
+        self.assertEqual(result["slides"][0]["issues"][0]["level"], "error")
        self.assertEqual(result["slides"][0]["issues"][0]["elements"], ["source"])

    def test_lint_xml_reports_text_out_of_canvas_and_warns_for_text_height(self) -> None:
@@ -660,8 +702,8 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            """
        )
        issue = result["slides"][0]["issues"][0]
-        self.assertEqual(result["summary"]["error_count"], 1)
-        self.assertEqual(result["summary"]["warning_count"], 1)
+        self.assertEqual(result["summary"]["error_count"], 2)
+        self.assertEqual(result["summary"]["warning_count"], 0)
        self.assertEqual(issue["code"], "shape_out_of_canvas")
        self.assertEqual(issue["overflow"], {"left": 0, "top": 0, "right": 160, "bottom": 40})

@@ -671,7 +713,7 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            <slide xmlns="http://www.larkoffice.com/sml/2.0">
              <data>
                <shape id="overflowing" type="text" topLeftX="80" topLeftY="80" width="360" height="80">
-                  <content fontSize="20" lineSpacing="multiple:1.5">
+                  <content fontSize="20" lineSpacing="multiple:1.5" autoFit="no-auto-fit">
                    <p>第一段</p><p>第二段</p><p>第三段</p><p>第四段</p>
                  </content>
                </shape>
@@ -685,20 +727,29 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
                    <p>第一段</p><p>第二段</p><p>第三段</p><p>第四段</p>
                  </content>
                </shape>
+                <shape id="shape-auto-fit" type="text" topLeftX="480" topLeftY="240" width="360" height="30">
+                  <content fontSize="20" lineSpacing="multiple:1.5" autoFit="shape-auto-fit">
+                    <p>第一段</p><p>第二段</p><p>第三段</p>
+                  </content>
+                </shape>
              </data>
            </slide>
            """
        )
        issues = result["slides"][0]["issues"]
-        self.assertEqual(result["summary"]["error_count"], 0)
-        self.assertEqual(result["summary"]["warning_count"], 1)
-        self.assertEqual(issues[0]["code"], "text_may_overflow_shape")
-        self.assertEqual(issues[0]["elements"], ["overflowing"])
-        self.assertEqual(issues[0]["line_count"], 4)
-        self.assertEqual(issues[0]["estimated_height"], 110)
-        self.assertEqual(issues[0]["available_height"], 80)
-        self.assertEqual(issues[0]["overflow"], 30)
-        self.assertIn('wrap="true" autoFit="normal-auto-fit"', issues[0]["message"])
+        overflow_issues = [issue for issue in issues if issue["code"] == "text_may_overflow_shape"]
+        self.assertEqual(result["summary"]["error_count"], 1)
+        overflow_ids = {issue["elements"][0] for issue in overflow_issues}
+        self.assertIn("overflowing", overflow_ids)
+        self.assertNotIn("auto-fit", overflow_ids)
+        self.assertNotIn("shape-auto-fit", overflow_ids)
+        self.assertNotIn("fitting", overflow_ids)
+        overflowing_issue = next(issue for issue in overflow_issues if issue["elements"] == ["overflowing"])
+        self.assertEqual(overflowing_issue["line_count"], 4)
+        self.assertEqual(overflowing_issue["estimated_height"], 110)
+        self.assertEqual(overflowing_issue["available_height"], 80)
+        self.assertEqual(overflowing_issue["overflow"], 30)
+        self.assertIn('wrap="true" autoFit="normal-auto-fit"', overflowing_issue["message"])

    def test_lint_xml_uses_fixed_line_spacing_for_text_height_warning(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
@@ -706,7 +757,7 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            <slide xmlns="http://www.larkoffice.com/sml/2.0">
              <data>
                <shape id="fixed-overflow" type="text" topLeftX="80" topLeftY="80" width="360" height="50">
-                  <content fontSize="20" lineSpacing="fixed:20">
+                  <content fontSize="20" lineSpacing="fixed:20" autoFit="no-auto-fit">
                    <p>第一段</p><p>第二段</p><p>第三段</p>
                  </content>
                </shape>
@@ -716,17 +767,525 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        )
        issue = result["slides"][0]["issues"][0]
        self.assertEqual(result["summary"]["warning_count"], 1)
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertEqual(issue["level"], "warning")
        self.assertEqual(issue["line_height"], 20)
        self.assertEqual(issue["estimated_height"], 60)
        self.assertEqual(issue["overflow"], 10)

+    def test_lint_xml_ignores_subpixel_text_height_overflow_tolerance(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="minor-overflow" type="text" topLeftX="80" topLeftY="80" width="360" height="39.8">
+                  <content fontSize="20" lineSpacing="fixed:20" autoFit="no-auto-fit">
+                    <p>第一段</p><p>第二段</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_allows_single_line_width_estimation_jitter(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="metric" type="text" topLeftX="80" topLeftY="80" width="152" height="54">
+                  <content fontSize="36" lineSpacing="multiple:1.2" autoFit="no-auto-fit"><p>4.16万亿</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_allows_short_metric_text_with_separators_as_single_line(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="metric" type="text" topLeftX="80" topLeftY="80" width="150" height="50">
+                  <content textType="title" fontSize="36" autoFit="no-auto-fit"><p>4.16万亿</p></content>
+                </shape>
+                <shape id="table-number" type="text" topLeftX="80" topLeftY="160" width="25" height="20">
+                  <content fontSize="10" textAlign="center" autoFit="no-auto-fit"><p>1,380</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_reports_plain_short_metric_when_it_wraps(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="plain-age" type="text" topLeftX="80" topLeftY="80" width="50" height="80">
+                  <content textType="title" fontSize="36" bold="true" autoFit="no-auto-fit"><p>82岁</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(len(overflow_issues), 1)
+        self.assertEqual(overflow_issues[0]["elements"], ["plain-age"])
+
+    def test_lint_xml_allows_centered_short_label_near_fit_as_single_line(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="centered-label" type="text" topLeftX="80" topLeftY="80" width="200" height="30">
+                  <content fontSize="14" bold="true" textAlign="center" autoFit="no-auto-fit">
+                    <p>参数服务器 (Parameter Server)</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_allows_headline_near_fit_as_single_line(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="headline" type="text" topLeftX="80" topLeftY="80" width="700" height="50">
+                  <content textType="headline" fontSize="26" bold="true" lineSpacing="multiple:1.3" autoFit="no-auto-fit">
+                    <p>全球半导体市场规模持续高速增长，AI驱动新一轮景气周期</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_allows_dense_body_line_spacing_estimation_slack(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="dense-body" type="text" topLeftX="80" topLeftY="80" width="360" height="140">
+                  <content fontSize="13" bold="true" lineSpacing="multiple:1.7" autoFit="no-auto-fit">
+                    <p>总体目标：</p>
+                    <p>建立深度神经网络高效训练的统一理论框架，实现训练效率与模型性能的协同优化。</p>
+                    <p>具体目标：</p>
+                    <p>提出自适应优化算法，收敛速度提升 2-3 倍</p>
+                    <p>实现结构化压缩方法，模型体积减少 10 倍以上</p>
+                    <p>构建分布式训练策略，64 GPU 加速比 &gt; 50x</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(overflow_issues, [])
+
+    def test_lint_xml_reports_dense_body_when_adjusted_height_still_overflows(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="dense-body" type="text" topLeftX="80" topLeftY="80" width="360" height="100">
+                  <content fontSize="13" bold="true" lineSpacing="multiple:1.7" autoFit="no-auto-fit">
+                    <p>总体目标：</p>
+                    <p>建立深度神经网络高效训练的统一理论框架，实现训练效率与模型性能的协同优化。</p>
+                    <p>具体目标：</p>
+                    <p>提出自适应优化算法，收敛速度提升 2-3 倍</p>
+                    <p>实现结构化压缩方法，模型体积减少 10 倍以上</p>
+                    <p>构建分布式训练策略，64 GPU 加速比 &gt; 50x</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(len(overflow_issues), 1)
+        self.assertEqual(overflow_issues[0]["elements"], ["dense-body"])
+
+    def test_lint_xml_reports_letter_spaced_caption_near_fit(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="caption" type="text" topLeftX="80" topLeftY="80" width="120" height="20">
+                  <content textType="caption" fontSize="11" letterSpacing="1" autoFit="no-auto-fit">
+                    <p>RISKS &amp; CHALLENGES</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(len(overflow_issues), 1)
+        self.assertEqual(overflow_issues[0]["elements"], ["caption"])
+
+    def test_lint_xml_reports_micro_caption_when_wrapping_overflows(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="micro-caption" type="text" topLeftX="80" topLeftY="60" width="200" height="16">
+                  <content textType="caption" fontSize="3" lineSpacing="multiple:1.3" letterSpacing="160" autoFit="no-auto-fit">
+                    <p>MARKET INSIGHT · 市场洞察</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        overflow_issues = [
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        ]
+        self.assertEqual(len(overflow_issues), 1)
+        self.assertEqual(overflow_issues[0]["elements"], ["micro-caption"])
+
+    def test_lint_xml_text_may_overflow_shape_upgrades_to_error_above_threshold(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="just-warning" type="text" topLeftX="80" topLeftY="80" width="360" height="50">
+                  <content fontSize="20" lineSpacing="fixed:20" autoFit="no-auto-fit">
+                    <p>第一段</p><p>第二段</p><p>第三段</p>
+                  </content>
+                </shape>
+                <shape id="error-overflow" type="text" topLeftX="80" topLeftY="200" width="360" height="30">
+                  <content fontSize="20" lineSpacing="fixed:20" autoFit="no-auto-fit">
+                    <p>第一段</p><p>第二段</p><p>第三段</p>
+                  </content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issues = {issue["elements"][0]: issue for issue in result["slides"][0]["issues"]}
+        self.assertEqual(issues["just-warning"]["level"], "warning")
+        self.assertEqual(issues["just-warning"]["overflow"], 10)
+        self.assertEqual(issues["error-overflow"]["level"], "error")
+        self.assertEqual(issues["error-overflow"]["overflow"], 30)
+        self.assertEqual(result["summary"]["error_count"], 1)
+        self.assertEqual(result["summary"]["warning_count"], 1)
+
+    def test_lint_xml_text_may_overflow_shape_downgrades_background_decoration_to_info(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="bg-deco" type="text" topLeftX="0" topLeftY="0" width="600" height="80" alpha="0.3">
+                  <content fontSize="120" lineSpacing="fixed:120" autoFit="no-auto-fit"><p>2026</p></content>
+                </shape>
+                <shape id="foreground" type="text" topLeftX="40" topLeftY="20" width="400" height="60">
+                  <content fontSize="20" lineSpacing="fixed:24"><p>Annual Report</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issues = {
+            issue["elements"][0]: issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape"
+        }
+        self.assertEqual(issues["bg-deco"]["level"], "info")
+        self.assertEqual(result["summary"]["warning_count"], 0)
+        self.assertEqual(result["summary"]["info_count"], 1)
+        self.assertEqual(result["slides"][0]["infos"], [issues["bg-deco"]])
+        self.assertIn("background decoration", issues["bg-deco"]["message"])
+
+    def test_lint_xml_allows_shape_alpha_ghost_text_out_of_canvas_and_overlap(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="ghost-number" type="text" topLeftX="-60" topLeftY="30" width="360" height="180" alpha="0.2">
+                  <content fontSize="160" lineSpacing="fixed:160" wrap="false"><p>01</p></content>
+                </shape>
+                <shape id="title" type="text" topLeftX="80" topLeftY="80" width="360" height="80">
+                  <content fontSize="30" lineSpacing="fixed:36"><p>Annual Review</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        codes = [issue["code"] for issue in result["slides"][0]["issues"]]
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertNotIn("shape_out_of_canvas", codes)
+        self.assertNotIn("bbox_overlap", codes)
+
+    def test_lint_xml_allows_content_color_alpha_ghost_text_out_of_canvas_and_overlap(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="ghost-year" type="text" topLeftX="760" topLeftY="20" width="260" height="160">
+                  <content fontSize="140" color="rgba(0,0,0,0.2)" lineSpacing="fixed:140" wrap="false"><p>2026</p></content>
+                </shape>
+                <shape id="headline" type="text" topLeftX="700" topLeftY="70" width="220" height="80">
+                  <content fontSize="28" lineSpacing="fixed:34"><p>Forecast</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        codes = [issue["code"] for issue in result["slides"][0]["issues"]]
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertNotIn("shape_out_of_canvas", codes)
+        self.assertNotIn("bbox_overlap", codes)
+
+    def test_lint_xml_allows_faint_medium_ghost_text_out_of_canvas_and_overlap(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="medium-ghost" type="text" topLeftX="820" topLeftY="300" width="270" height="72" alpha="0.32">
+                  <content fontSize="40" lineSpacing="fixed:40" wrap="false"><p>OFF EDGE</p></content>
+                </shape>
+                <shape id="caption" type="text" topLeftX="760" topLeftY="315" width="180" height="36">
+                  <content fontSize="16" lineSpacing="fixed:20"><p>Readable caption</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        codes = [issue["code"] for issue in result["slides"][0]["issues"]]
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertNotIn("shape_out_of_canvas", codes)
+        self.assertNotIn("bbox_overlap", codes)
+
+    def test_lint_xml_allows_ghost_text_image_overlap(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="ghost-label" type="text" topLeftX="100" topLeftY="40" width="560" height="160" alpha="0.2">
+                  <content fontSize="120" lineSpacing="fixed:120" wrap="false"><p>2026</p></content>
+                </shape>
+                <img id="photo" src="token" topLeftX="160" topLeftY="70" width="260" height="160"/>
+                <shape id="title" type="text" topLeftX="610" topLeftY="95" width="320" height="60">
+                  <content fontSize="28" lineSpacing="fixed:34"><p>Annual Review</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        codes = [issue["code"] for issue in result["slides"][0]["issues"]]
+        self.assertNotIn("image_covers_text", codes)
+        self.assertNotIn("bbox_overlap", codes)
+
+    def test_lint_slide_allows_ghost_text_whiteboard_overlap(self) -> None:
+        result = xml_text_overlap_lint.lint_slide(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <whiteboard id="board" topLeftX="180" topLeftY="70" width="420" height="300"/>
+                <shape id="ghost-label" type="text" topLeftX="100" topLeftY="40" width="560" height="160" alpha="0.2">
+                  <content fontSize="120" lineSpacing="fixed:120" wrap="false"><p>2026</p></content>
+                </shape>
+                <shape id="title" type="text" topLeftX="610" topLeftY="95" width="220" height="60">
+                  <content fontSize="28" lineSpacing="fixed:34"><p>Annual Review</p></content>
+                </shape>
+              </data>
+            </slide>
+            """,
+            1,
+        )
+        codes = [issue["code"] for issue in result["issues"]]
+        self.assertNotIn("whiteboard_external_overlap", codes)
+
+    def test_lint_xml_allows_faint_ghost_text_without_area_threshold(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="small-ghost" type="text" topLeftX="940" topLeftY="300" width="40" height="40" alpha="0.32">
+                  <content fontSize="36" lineSpacing="fixed:36" wrap="false"><p>土</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertEqual(result["slides"][0]["issues"], [])
+
+    def test_lint_xml_keeps_out_of_canvas_error_for_medium_text_without_faint_alpha(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="medium-not-ghost" type="text" topLeftX="820" topLeftY="300" width="270" height="72" alpha="0.36">
+                  <content fontSize="54" lineSpacing="fixed:54" wrap="false"><p>OFF EDGE</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        self.assertEqual(result["summary"]["error_count"], 1)
+        self.assertEqual(result["slides"][0]["issues"][0]["code"], "shape_out_of_canvas")
+        self.assertEqual(result["slides"][0]["issues"][0]["elements"], ["medium-not-ghost"])
+
+    def test_lint_xml_keeps_out_of_canvas_error_for_half_alpha_large_text(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="half-alpha" type="text" topLeftX="760" topLeftY="20" width="260" height="160">
+                  <content fontSize="140" color="rgba(0,0,0,0.5)" lineSpacing="fixed:140" wrap="false"><p>2026</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        self.assertEqual(result["summary"]["error_count"], 1)
+        self.assertEqual(result["slides"][0]["issues"][0]["code"], "shape_out_of_canvas")
+        self.assertEqual(result["slides"][0]["issues"][0]["elements"], ["half-alpha"])
+
+    def test_lint_xml_text_may_overflow_shape_keeps_error_when_alpha_not_low(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="opaque-big" type="text" topLeftX="0" topLeftY="0" width="600" height="80" alpha="0.9">
+                  <content fontSize="120" lineSpacing="fixed:120" autoFit="no-auto-fit"><p>2026</p></content>
+                </shape>
+                <shape id="foreground" type="text" topLeftX="40" topLeftY="20" width="400" height="60">
+                  <content fontSize="20" lineSpacing="fixed:24"><p>Annual Report</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issue = next(
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape" and issue["elements"] == ["opaque-big"]
+        )
+        self.assertEqual(issue["level"], "error")
+
+    def test_lint_xml_text_may_overflow_shape_keeps_error_when_no_foreground_text(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="lonely-big" type="text" topLeftX="0" topLeftY="0" width="600" height="80" alpha="0.3">
+                  <content fontSize="120" lineSpacing="fixed:120" autoFit="no-auto-fit"><p>2026</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issue = next(
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape" and issue["elements"] == ["lonely-big"]
+        )
+        self.assertEqual(issue["level"], "error")
+
+    def test_lint_xml_text_may_overflow_shape_keeps_error_when_foreground_alpha_zero(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="bg-deco" type="text" topLeftX="0" topLeftY="0" width="600" height="80" alpha="0.3">
+                  <content fontSize="120" lineSpacing="fixed:120" autoFit="no-auto-fit"><p>2026</p></content>
+                </shape>
+                <shape id="transparent-foreground" type="text" topLeftX="40" topLeftY="20" width="400" height="60" alpha="0">
+                  <content fontSize="20" lineSpacing="fixed:24"><p>Annual Report</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issue = next(
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape" and issue["elements"] == ["bg-deco"]
+        )
+        self.assertEqual(issue["level"], "error")
+
+    def test_lint_xml_text_may_overflow_shape_keeps_error_when_foreground_is_below_in_order(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="foreground" type="text" topLeftX="40" topLeftY="20" width="400" height="60">
+                  <content fontSize="20" lineSpacing="fixed:24"><p>Annual Report</p></content>
+                </shape>
+                <shape id="top-big" type="text" topLeftX="0" topLeftY="0" width="600" height="80" alpha="0.3">
+                  <content fontSize="120" lineSpacing="fixed:120" autoFit="no-auto-fit"><p>2026</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issue = next(
+            issue
+            for issue in result["slides"][0]["issues"]
+            if issue["code"] == "text_may_overflow_shape" and issue["elements"] == ["top-big"]
+        )
+        self.assertEqual(issue["level"], "error")
+
    def test_lint_xml_uses_paragraph_spacing_overrides_for_text_height_warning(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <slide xmlns="http://www.larkoffice.com/sml/2.0">
              <data>
-                <shape id="paragraph-overflow" type="text" topLeftX="80" topLeftY="80" width="360" height="35">
-                  <content fontSize="20" lineSpacing="multiple:1.5">
+                <shape id="paragraph-overflow" type="text" topLeftX="80" topLeftY="80" width="360" height="30">
+                  <content fontSize="20" lineSpacing="multiple:1.5" autoFit="no-auto-fit">
                    <p lineSpacing="fixed:10" beforeLineSpacing="fixed:5" afterLineSpacing="fixed:5">第一行<br/>第二行</p>
                  </content>
                </shape>
@@ -745,7 +1304,36 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        self.assertEqual(issues[0]["line_count"], 2)
        self.assertEqual(issues[0]["line_height"], 10)
        self.assertEqual(issues[0]["estimated_height"], 40)
-        self.assertEqual(issues[0]["overflow"], 5)
+        self.assertEqual(issues[0]["overflow"], 10)
+
+    def test_lint_xml_uses_letter_spacing_for_text_overflow_warning(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <shape id="baseline" type="text" topLeftX="0" topLeftY="0" width="120" height="30">
+                  <content fontSize="20" lineSpacing="multiple:1.5" autoFit="no-auto-fit"><p>一二三四五六</p></content>
+                </shape>
+                <shape id="content-spaced" type="text" topLeftX="200" topLeftY="0" width="120" height="30">
+                  <content fontSize="20" lineSpacing="multiple:1.5" letterSpacing="2" autoFit="no-auto-fit"><p>一二三四五六</p></content>
+                </shape>
+                <shape id="paragraph-spaced" type="text" topLeftX="400" topLeftY="0" width="120" height="30">
+                  <content fontSize="20" lineSpacing="multiple:1.5" autoFit="no-auto-fit"><p letterSpacing="2">一二三四五六</p></content>
+                </shape>
+              </data>
+            </slide>
+            """
+        )
+        issues = result["slides"][0]["issues"]
+        overflow_ids = [issue["elements"][0] for issue in issues if issue["code"] == "text_may_overflow_shape"]
+        self.assertNotIn("baseline", overflow_ids)
+        self.assertIn("content-spaced", overflow_ids)
+        self.assertIn("paragraph-spaced", overflow_ids)
+        by_id = {issue["elements"][0]: issue for issue in issues if issue["code"] == "text_may_overflow_shape"}
+        self.assertEqual(by_id["content-spaced"]["line_count"], 2)
+        self.assertEqual(by_id["content-spaced"]["estimated_height"], 50)
+        self.assertEqual(by_id["content-spaced"]["overflow"], 20)
+        self.assertEqual(by_id["paragraph-spaced"]["line_count"], 2)

    def test_strip_xml_paragraphs_preserves_br_as_hard_line_break(self) -> None:
        self.assertEqual(
@@ -753,7 +1341,7 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            "第一行\n第二行\n第三行",
        )

-    def test_lint_xml_blocks_template_style_bleed_outside_canvas(self) -> None:
+    def test_lint_xml_allows_template_style_images_outside_canvas(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
@@ -771,9 +1359,8 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </presentation>
            """
        )
-        self.assertEqual(result["summary"]["error_count"], 1)
+        self.assertEqual(result["summary"]["error_count"], 0)
        self.assertEqual(result["summary"]["warning_count"], 0)
-        self.assertEqual(result["slides"][0]["errors"][0]["code"], "img_out_of_canvas")

    def test_extract_elements_preserves_supported_element_geometry_order_and_text_metadata(self) -> None:
        elements = xml_text_overlap_lint.extract_elements(
@@ -790,6 +1377,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
                <table id="table" topLeftX="400" topLeftY="60" width="220" height="120"></table>
                <chart id="chart" topLeftX="640" topLeftY="60" width="220" height="120"/>
                <whiteboard id="wb" topLeftX="80" topLeftY="220" width="760" height="240"/>
+                <embed id="emb" topLeftX="600" topLeftY="320" width="240" height="140">
+                  <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 240 140"><rect x="0" y="0" width="240" height="140"/></svg>
+                </embed>
                <shape id="missing-height" type="text" topLeftX="80" topLeftY="480" width="320">
                  <content><p>Skipped</p></content>
                </shape>
@@ -797,9 +1387,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </slide>
            """
        )
-        self.assertEqual([element["id"] for element in elements], ["photo", "headline", "table", "chart", "wb"])
-        self.assertEqual([element["kind"] for element in elements], ["img", "shape", "table", "chart", "whiteboard"])
-        self.assertEqual([element["order"] for element in elements], [0, 1, 2, 3, 4])
+        self.assertEqual([element["id"] for element in elements], ["photo", "headline", "table", "chart", "wb", "emb"])
+        self.assertEqual([element["kind"] for element in elements], ["img", "shape", "table", "chart", "whiteboard", "embed"])
+        self.assertEqual([element["order"] for element in elements], [0, 1, 2, 3, 4, 5])
        self.assertEqual(elements[1]["type"], "text")
        self.assertEqual(elements[1]["textType"], "headline")
        self.assertEqual(elements[1]["textAlign"], "center")
@@ -807,7 +1397,7 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        self.assertEqual(elements[1]["fontSize"], 28)
        self.assertEqual(elements[1]["text"], "Growth & scale\nFocused execution")

-    def test_lint_xml_blocks_small_out_of_bounds_images(self) -> None:
+    def test_lint_xml_ignores_small_out_of_bounds_images(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
@@ -819,10 +1409,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </presentation>
            """
        )
-        self.assertEqual(result["summary"]["error_count"], 1)
-        self.assertEqual(result["slides"][0]["errors"][0]["code"], "img_out_of_canvas")
+        self.assertEqual(result["summary"]["error_count"], 0)

-    def test_lint_xml_blocks_out_of_canvas_images(self) -> None:
+    def test_lint_xml_ignores_out_of_canvas_images(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
@@ -835,13 +1424,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </presentation>
            """
        )
-        self.assertEqual(result["summary"]["error_count"], 2)
-        self.assertEqual(
-            [issue["code"] for issue in result["slides"][0]["errors"]],
-            ["img_out_of_canvas", "img_out_of_canvas"],
-        )
+        self.assertEqual(result["summary"]["error_count"], 0)

-    def test_lint_xml_blocks_full_bleed_images_outside_canvas(self) -> None:
+    def test_lint_xml_ignores_full_bleed_images_outside_canvas(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
@@ -853,10 +1438,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            </presentation>
            """
        )
-        self.assertEqual(result["summary"]["error_count"], 1)
-        self.assertEqual(result["slides"][0]["errors"][0]["code"], "img_out_of_canvas")
+        self.assertEqual(result["summary"]["error_count"], 0)

-    def test_lint_xml_reports_text_and_chart_out_of_canvas(self) -> None:
+    def test_lint_xml_reports_text_and_chart_but_not_image_out_of_canvas(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
@@ -871,17 +1455,16 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            """
        )
        issues = result["slides"][0]["issues"]
-        self.assertEqual(result["summary"]["error_count"], 3)
+        self.assertEqual(result["summary"]["error_count"], 2)
        self.assertEqual(
            [(issue["code"], issue["elements"], issue["overflow"]) for issue in issues],
            [
                ("shape_out_of_canvas", ["outside-shape"], {"left": 10, "top": 0, "right": 0, "bottom": 0}),
-                ("img_out_of_canvas", ["outside-img"], {"left": 0, "top": 20, "right": 0, "bottom": 0}),
                ("chart_out_of_canvas", ["outside-chart"], {"left": 0, "top": 0, "right": 40, "bottom": 0}),
            ],
        )

-    def test_lint_xml_reports_line_out_of_canvas_with_structured_geometry(self) -> None:
+    def test_lint_xml_ignores_line_out_of_canvas(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <slide xmlns="http://www.larkoffice.com/sml/2.0">
@@ -895,11 +1478,8 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            """
        )

-        issue = result["slides"][0]["errors"][0]
-        self.assertEqual(issue["code"], "line_out_of_canvas")
-        self.assertEqual(issue["element_ids"], ["connector"])
-        self.assertEqual(issue["measurement"]["overflow"]["right"], 20)
-        self.assertEqual(issue["related_objects"][0]["kind"], "line")
+        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertEqual(result["slides"][0]["issues"], [])

    def test_lint_xml_uses_rotated_text_and_chart_bounds_for_canvas_validation(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
@@ -922,13 +1502,13 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        self.assertEqual(issues_by_element["rotated-chart"]["code"], "chart_out_of_canvas")
        self.assertAlmostEqual(issues_by_element["rotated-chart"]["overflow"]["right"], 20.710678, places=5)

-    def test_lint_xml_uses_rotated_bounds_for_rect_and_image_canvas_validation(self) -> None:
+    def test_lint_xml_uses_declared_bounds_for_rect_and_ignores_images(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
            <presentation xmlns="http://www.larkoffice.com/sml/2.0" width="960" height="540">
              <slide xmlns="http://www.larkoffice.com/sml/2.0">
                <data>
-                  <shape id="rotated-rect" type="rect" topLeftX="0" topLeftY="0" width="100" height="100" rotation="45"/>
+                  <shape id="rotated-rect" type="rect" topLeftX="900" topLeftY="0" width="100" height="100" rotation="45"/>
                  <img id="rotated-image" topLeftX="860" topLeftY="200" width="100" height="100" rotation="45"/>
                </data>
              </slide>
@@ -936,12 +1516,54 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            """
        )
        issues_by_element = {issue["elements"][0]: issue for issue in result["slides"][0]["issues"]}
-        self.assertEqual(result["summary"]["error_count"], 2)
+        self.assertEqual(result["summary"]["error_count"], 1)
        self.assertEqual(issues_by_element["rotated-rect"]["code"], "shape_out_of_canvas")
-        self.assertAlmostEqual(issues_by_element["rotated-rect"]["overflow"]["left"], 20.710678, places=5)
-        self.assertAlmostEqual(issues_by_element["rotated-rect"]["overflow"]["top"], 20.710678, places=5)
-        self.assertEqual(issues_by_element["rotated-image"]["code"], "img_out_of_canvas")
-        self.assertAlmostEqual(issues_by_element["rotated-image"]["overflow"]["right"], 20.710678, places=5)
+        self.assertEqual(issues_by_element["rotated-rect"]["overflow"], {"left": 0, "top": 0, "right": 40, "bottom": 0})
+        self.assertNotIn("rotated-image", issues_by_element)
+
+    def test_detect_elements_out_of_canvas_limits_detection_to_whitelist(self) -> None:
+        issues = xml_text_overlap_lint.detect_elements_out_of_canvas(
+            [
+                {"id": "table", "kind": "table", "x": 95, "y": 0, "width": 10, "height": 10, "rotation": 45},
+                {"id": "chart", "kind": "chart", "x": 95, "y": 0, "width": 10, "height": 10, "rotation": 0},
+                {
+                    "id": "text",
+                    "kind": "shape",
+                    "type": "text",
+                    "x": 95,
+                    "y": 0,
+                    "width": 10,
+                    "height": 10,
+                    "rotation": 0,
+                },
+                {
+                    "id": "rect",
+                    "kind": "shape",
+                    "type": "rect",
+                    "x": 95,
+                    "y": 0,
+                    "width": 10,
+                    "height": 10,
+                    "rotation": 45,
+                },
+                {"id": "image", "kind": "img", "x": 95, "y": 0, "width": 10, "height": 10, "rotation": 0},
+                {
+                    "id": "ellipse",
+                    "kind": "shape",
+                    "type": "ellipse",
+                    "x": 95,
+                    "y": 0,
+                    "width": 10,
+                    "height": 10,
+                    "rotation": 0,
+                },
+            ],
+            100,
+            100,
+        )
+
+        self.assertEqual([issue["elements"] for issue in issues], [["table"], ["chart"], ["text"], ["rect"]])
+        self.assertEqual(issues[-1]["bbox"], {"x": 95, "y": 0, "width": 10, "height": 10})

    def test_lint_xml_treats_non_finite_rotations_as_zero(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
@@ -1107,8 +1729,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        )
        issues_by_dimension = {issue["dimension"]: issue for issue in result["slides"][0]["issues"]}
        self.assertEqual(result["summary"]["error_count"], 0)
-        self.assertEqual(result["summary"]["warning_count"], 2)
-        self.assertEqual(issues_by_dimension["width"]["level"], "warning")
+        self.assertEqual(result["summary"]["warning_count"], 0)
+        self.assertEqual(result["summary"]["info_count"], 2)
+        self.assertEqual(issues_by_dimension["width"]["level"], "info")
        self.assertEqual(issues_by_dimension["width"]["code"], "table_resolved_size_mismatch")
        self.assertEqual(issues_by_dimension["width"]["resolved_sizes"], [100, 100, 50])
        self.assertEqual(issues_by_dimension["width"]["resolved_size"], 250)
@@ -1201,7 +1824,7 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
        }
        script_path = Path(xml_text_overlap_lint.__file__).resolve()
        with tempfile.TemporaryDirectory() as temp_dir:
-            for name, (table_xml, expected_warning_count) in cases.items():
+            for name, (table_xml, expected_info_count) in cases.items():
                with self.subTest(case=name):
                    input_path = Path(temp_dir) / f"{name}.xml"
                    input_path.write_text(
@@ -1221,9 +1844,10 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
                    result = json.loads(completed.stdout)
                    self.assertEqual(completed.returncode, 0, completed.stderr)
                    self.assertEqual(result["summary"]["error_count"], 0)
-                    self.assertEqual(result["summary"]["warning_count"], expected_warning_count)
+                    self.assertEqual(result["summary"]["warning_count"], 0)
+                    self.assertEqual(result["summary"]["info_count"], expected_info_count)
                    self.assertTrue(
-                        all(issue["level"] == "warning" for issue in result["slides"][0]["issues"]),
+                        all(issue["level"] == "info" for issue in result["slides"][0]["issues"]),
                        result["slides"][0]["issues"],
                    )

@@ -1268,8 +1892,9 @@ class XmlTextOverlapLintGeometryTest(unittest.TestCase):
            """
        )
        issue = next(issue for issue in result["slides"][0]["issues"] if issue["code"] == "image_may_cover_vertical_text")
-        self.assertEqual(issue["level"], "warning")
+        self.assertEqual(issue["level"], "info")
        self.assertEqual(result["summary"]["error_count"], 0)
+        self.assertEqual(result["summary"]["info_count"], 1)


 class XmlTextOverlapLintDensityTest(unittest.TestCase):
@@ -1357,7 +1982,7 @@ class XmlTextOverlapLintDensityTest(unittest.TestCase):
            "id": "sparse_container_content",
        })
        self.assertEqual(issue["measurement"]["container_area"], 151700)
-        self.assertEqual(issue["measurement"]["content_coverage_ratio"], 0.032)
+        self.assertEqual(issue["measurement"]["content_coverage_ratio"], 0.03)
        self.assertEqual(issue["elements"], ["trend-card", "trend-title", "trend-copy"])
        self.assertEqual(issue["element_ids"], ["trend-card", "trend-title", "trend-copy"])
        self.assertEqual(
@@ -1803,6 +2428,25 @@ class XmlTextOverlapLintDensityTest(unittest.TestCase):

        self.assertEqual(result["slides"][0]["issues"], [])

+    def test_lint_xml_does_not_report_blank_slide_for_embed_only_content(self) -> None:
+        result = xml_text_overlap_lint.lint_xml(
+            """
+            <slide xmlns="http://www.larkoffice.com/sml/2.0">
+              <data>
+                <embed id="emb" topLeftX="280" topLeftY="130" width="400" height="280">
+                  <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 400 280">
+                    <circle cx="200" cy="140" r="100" fill="#2563EB"/>
+                  </svg>
+                </embed>
+              </data>
+            </slide>
+            """
+        )
+
+        self.assertEqual(result["summary"]["error_count"], 0)
+        codes = [issue["code"] for issue in result["slides"][0]["issues"]]
+        self.assertNotIn("blank_slide", codes)
+
    def test_lint_xml_does_not_report_blank_slide_for_line_only_content(self) -> None:
        result = xml_text_overlap_lint.lint_xml(
            """
@@ -1842,9 +2486,9 @@ class XmlTextOverlapLintDensityTest(unittest.TestCase):
        # Must match the visual bbox that should_flag_overlap actually decided with (fontSize=14
        # from extract_elements), not the fontSize=96 max-descendant value that
        # extract_density_elements computes for the same "left" element id.
-        self.assertEqual(issue["measurement"]["intersection_width"], 117.04)
+        self.assertEqual(issue["measurement"]["intersection_width"], 109.2)
        self.assertEqual(issue["measurement"]["intersection_height"], 6.8)
-        self.assertEqual(issue["measurement"]["intersection_area"], 795.872)
+        self.assertEqual(issue["measurement"]["intersection_area"], 742.56)

    def test_has_similar_short_card_peer_excludes_the_element_itself(self) -> None:
        card_a = {"kind": "shape", "type": "rect", "x": 0, "y": 0, "width": 300, "height": 100}
--- a/skills/lark-task/references/lark-task-create.md
+++ b/skills/lark-task/references/lark-task-create.md
@@ -24,6 +24,12 @@ lark-cli task +create \
 lark-cli task +create \
  --summary "Buy milk"

+# Create a milestone by passing an API field without a named flag
+lark-cli task +create \
+  --summary "Release v2.0" \
+  --due "2026-08-15" \
+  --data '{"is_milestone":true}'
+
 # Preview the API call without executing
 lark-cli task +create --summary "Test Task" --dry-run
 ```
@@ -39,8 +45,11 @@ lark-cli task +create --summary "Test Task" --dry-run
 | `--due <time>` | No | Due date. Supports ISO 8601, `YYYY-MM-DD`, relative time (e.g., `+2d`), or ms timestamp. `YYYY-MM-DD` and relative time will automatically set it as an all-day task. |
 | `--tasklist-id <id>` | No | The GUID of the tasklist, or a full AppLink URL (the CLI will automatically extract the `guid` parameter from the URL). |
 | `--idempotency-key <key>` | No | Client token to ensure idempotency of the request. |
+| `--data <json>` | No | JSON object merged into the task create request for API fields without dedicated flags, such as `{"is_milestone":true}`. Explicit named flags override same-named fields in this object. |
 | `--dry-run` | No | Preview the API call (JSON payload) without actually creating the task. |

+Use `lark-cli schema task.tasks.create` to confirm that an extra field is supported before passing it through `--data`. Prefer this shortcut over the raw `tasks create` command when `--data` can express the request. Do not assume that other shortcuts support `--data`; check each shortcut's `--help` output first.
+
 ## Workflow

 1. Confirm with the user: task summary, due date, assignee, and tasklist if necessary.
--- a/tests/cli_e2e/base/base_role_workflow_test.go
+++ b/tests/cli_e2e/base/base_role_workflow_test.go
@@ -111,18 +111,29 @@ func TestBase_RoleWorkflow(t *testing.T) {
 		result.AssertExitCode(t, 0)
 		result.AssertStdoutStatus(t, true)

-		getResult, err := clie2e.RunCmd(ctx, clie2e.Request{
-			Args:      []string{"base", "+role-get", "--base-token", baseToken, "--role-id", roleID},
-			DefaultAs: "bot",
-		})
-		require.NoError(t, err)
-		getResult.AssertExitCode(t, 0)
-		getResult.AssertStdoutStatus(t, true)
+		pollTimeout := 30 * time.Second
+		pollCtx, pollCancel := context.WithTimeout(ctx, pollTimeout)
+		defer pollCancel()

-		rolePayload := gjson.Get(getResult.Stdout, "data.data").String()
-		require.NotEmpty(t, rolePayload, "stdout:\n%s", getResult.Stdout)
-		require.True(t, gjson.Valid(rolePayload), "stdout:\n%s", getResult.Stdout)
-		assert.Equal(t, updatedRoleName, gjson.Get(rolePayload, "role_name").String())
+		err = clie2e.WaitForCondition(pollCtx, clie2e.WaitOptions{
+			Timeout:  pollTimeout,
+			Interval: 3 * time.Second,
+		}, func() (bool, error) {
+			getResult, getErr := clie2e.RunCmd(pollCtx, clie2e.Request{
+				Args:      []string{"base", "+role-get", "--base-token", baseToken, "--role-id", roleID},
+				DefaultAs: "bot",
+			})
+			if getErr != nil {
+				return false, getErr
+			}
+			if getResult.ExitCode != 0 {
+				return false, getResult.RunErr
+			}
+
+			rolePayload := gjson.Get(getResult.Stdout, "data.data").String()
+			return gjson.Valid(rolePayload) && gjson.Get(rolePayload, "role_name").String() == updatedRoleName, nil
+		})
+		require.NoError(t, err, "role name should converge to %q", updatedRoleName)
 	})

 }
--- a/tests/cli_e2e/calendar/calendar_description_rich_dryrun_test.go
+++ b/tests/cli_e2e/calendar/calendar_description_rich_dryrun_test.go
@@ -17,10 +17,9 @@ import (
 // converts Markdown <-> ClientVars.
 const richTextMarkdown = "见 [设计文档](https://bytedance.feishu.cn/docx/abc) 和 **重点**"

-// TestCalendar_CreateDescriptionRichDryRun verifies that +create forwards the
-// rich-text payload as-is under the description_rich body field, and that even
-// when the deprecated --description is also supplied, the CLI sends only
-// description_rich (rich wins) and omits the plain description field — the
+// TestCalendar_CreateDescriptionRichDryRun verifies that +create treats
+// --description as Markdown rich text and forwards it as-is under the
+// description_rich body field, omitting the plain description field — the
 // service treats the two as mutually exclusive.
 func TestCalendar_CreateDescriptionRichDryRun(t *testing.T) {
 	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
@@ -38,8 +37,7 @@ func TestCalendar_CreateDescriptionRichDryRun(t *testing.T) {
 			"--summary", "rich dry-run",
 			"--start", "2026-04-25T10:00:00+08:00",
 			"--end", "2026-04-25T11:00:00+08:00",
-			"--description", "plain fallback",
-			"--description-rich", richTextMarkdown,
+			"--description", richTextMarkdown,
 			"--dry-run",
 		},
 		DefaultAs: "bot",
@@ -54,10 +52,10 @@ func TestCalendar_CreateDescriptionRichDryRun(t *testing.T) {
 	require.False(t, clie2e.DryRunGet(out, "api.0.body.description").Exists(), "plain description must not be sent; stdout:\n%s", out)
 }

-// TestCalendar_CreateDescriptionRichOnlyDryRun verifies that when only
-// --description-rich is provided (no --description), +create omits the
-// description body field entirely. Sending an empty description would suppress
-// the server's plain-preview backfill and break first-load rendering.
+// TestCalendar_CreateDescriptionRichOnlyDryRun verifies that +create forwards
+// --description under description_rich and omits the plain description body
+// field entirely. Sending an empty description would suppress the server's
+// plain-preview backfill and break first-load rendering.
 func TestCalendar_CreateDescriptionRichOnlyDryRun(t *testing.T) {
 	t.Setenv("LARKSUITE_CLI_CONFIG_DIR", t.TempDir())
 	t.Setenv("LARKSUITE_CLI_APP_ID", "app")
@@ -74,7 +72,7 @@ func TestCalendar_CreateDescriptionRichOnlyDryRun(t *testing.T) {
 			"--summary", "rich only dry-run",
 			"--start", "2026-04-25T10:00:00+08:00",
 			"--end", "2026-04-25T11:00:00+08:00",
-			"--description-rich", richTextMarkdown,
+			"--description", richTextMarkdown,
 			"--dry-run",
 		},
 		DefaultAs: "bot",
@@ -103,7 +101,7 @@ func TestCalendar_UpdateDescriptionRichDryRun(t *testing.T) {
 			"calendar", "+update",
 			"--calendar-id", "cal_dry",
 			"--event-id", "evt_dry",
-			"--description-rich", richTextMarkdown,
+			"--description", richTextMarkdown,
 			"--notify=false",
 			"--dry-run",
 		},
--- a/tests/cli_e2e/sheets/sheets_image_upload_dryrun_test.go
+++ b/tests/cli_e2e/sheets/sheets_image_upload_dryrun_test.go
@@ -17,11 +17,10 @@ import (
 // TestSheets_ImageUploadDryRunParentType pins the parent_type the sheets
 // image-upload shortcuts emit in --dry-run output for native vs. imported
 // "office" spreadsheets. For native tokens parent_type must be "sheet_image";
-// for tokens prefixed with "fake_office_" (the synthetic token an imported
-// office spreadsheet carries) the backend requires "office_sheet_file". The
-// three covered entries — sheets +media-upload (backward), sheets
-// +cells-set-image, and sheets +create-float-image — are every image-upload
-// surface that the office/native split fans out to.
+// for tokens carrying the interleaved "OFL0X" marker the backend requires
+// "office_sheet_file". The covered entries — sheets +media-upload (backward),
+// sheets +cells-set-image, and sheets +float-image-create — are every
+// image-upload surface that the office/native split fans out to.
 func TestSheets_ImageUploadDryRunParentType(t *testing.T) {
 	setSheetsDryRunEnv(t)

@@ -50,11 +49,11 @@ func TestSheets_ImageUploadDryRunParentType(t *testing.T) {
 			name: "media-upload office",
 			args: []string{
 				"sheets", "+media-upload",
-				"--spreadsheet-token", "fake_office_dryrun",
+				"--spreadsheet-token", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
 				"--file", "img.png",
 				"--dry-run",
 			},
-			token:          "fake_office_dryrun",
+			token:          "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
 			wantParentType: "office_sheet_file",
 		},
 		{
@@ -74,13 +73,30 @@ func TestSheets_ImageUploadDryRunParentType(t *testing.T) {
 			name: "cells-set-image office",
 			args: []string{
 				"sheets", "+cells-set-image",
-				"--spreadsheet-token", "fake_office_dryrun",
+				"--spreadsheet-token", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
 				"--sheet-id", "sheet1",
 				"--range", "A1",
 				"--image", "img.png",
 				"--dry-run",
 			},
-			token:          "fake_office_dryrun",
+			token:          "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
+			wantParentType: "office_sheet_file",
+		},
+		{
+			name: "float-image-create office",
+			args: []string{
+				"sheets", "+float-image-create",
+				"--spreadsheet-token", "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
+				"--sheet-id", "sheet1",
+				"--image-name", "img.png",
+				"--image", "img.png",
+				"--position-row", "0",
+				"--position-col", "A",
+				"--size-width", "100",
+				"--size-height", "100",
+				"--dry-run",
+			},
+			token:          "aaaaOaaaaFaaaaLaaaa0aaaaXaaa",
 			wantParentType: "office_sheet_file",
 		},
 	}
Author	SHA1	Message	Date
niezhiwei	f2b77acfdb	fix(slides): lint 放行并识别 embed 嵌入元素 schema 已支持 <embed> 嵌入 SVG 协议,但版式准出 lint 未同步: - SXSD 标签校验仅在 <whiteboard> 祖先下放行 SVG 命名空间子树, <embed> 内的合法 <svg> 会被判 sxsd_unsupported_tag 并 block - extract_elements 不提取 embed,仅含 embed 的页面被误报 blank_slide,也不参与重叠/密度检查修复:SVG 子树白名单加入 embed 祖先并更新 svg 报错 hint;embed 进入元素提取以及 slide_content_visual_bbox、is_large_visual_child 的视觉元素集合。新增 embed 内 svg 放行、embed-only 不报空白页两个测试,extract_elements 测试覆盖 embed。 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>	2026-07-29 11:59:13 +08:00
niezhiwei	91edc443e8	feat(slides): 支持嵌入 SVG 内容协议同步 ai_xsd MR 62（feat-vector-svg-protocol）到 CLI 的 SML schema 副本： - schema 新增 embed 元素：嵌入内容容器，外层承载 Slides 摆放和效果属性（topLeftX/topLeftY/width/height 必填，rotation/flipX/flipY/alpha 可选），内层为标准 SVG（http://www.w3.org/2000/svg 命名空间），可选 reflection/shadow 子元素 - data 容器 choice 列表注册 sml:embed - xml-schema-quick-ref.md 同步 data 元素列表并新增 embed 小节与示例 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>	2026-07-29 11:30:29 +08:00
R0bynZhu	358cd06838	docs(slides): 补齐 shortcut 参数说明，修正 +xml-get --output 必填标注 (#2088 ) * docs(slides): consolidate CWD-relative path rule into one global rule State the "all local file path args must be CWD-relative (absolute rejected)" rule once in SKILL.md 权威经验, and trim the per-command repetitions in media-upload / create / screenshot / xml-presentations-get. Also fix the stale xml-presentations-get param table: --output is optional (relative), not required. * feat: try common solution * chore: 优化措辞 * feat: 优化措辞 * feat: 优化措辞 * docs(slides): 强调调用命令前必读对应命令文档 - 「调用命令前再读」改为「调用相关命令前必须读取相关的文档以了解命令的使用方式」，并把原「按需再读」列表合并进来，去掉可选语义 - 移除 lark-shared 的 CRITICAL 前置阅读要求 - Step 4 回读示例补全 `--presentation <xml_presentation_id>` 参数 * docs(slides): Shortcuts 表补充 +screenshot 并写明本地路径参数 - 新增 +screenshot 行：--slide-number 页号（从 1 开始，可重复，一次最多 10 页）、 --output-dir 保存目录（CWD 内相对路径，默认 .lark-slides/screenshots） - +xml-get 行补上 --presentation 和 --output（CWD 内相对路径），并说明省略 --output 时 XML 返回在 JSON 信封里 * revert(slides): 回退 references 下的文档改动，只保留 SKILL.md 把 lark-slides-create.md、lark-slides-media-upload.md、lark-slides-screenshot.md、 lark-slides-xml-presentations-get.md 还原为 main 的版本，本分支只改 SKILL.md。 * docs(slides): 恢复开始前必读 lark-shared 的 CRITICAL 要求认证、权限和全局参数以 lark-shared 为准，这条前置阅读不该在本分支被删掉。 * chore: 移除output省略的说明	2026-07-29 10:55:05 +08:00
Yuxuan Zhao	b0b1ca4b5d	test(e2e): wait for base role update visibility (#2087 )	2026-07-28 21:45:00 +08:00
liangshuo-1	781d188a60	chore: release v1.0.79 (#2082 )	2026-07-28 21:02:37 +08:00
calendar-assistant	2e0fb9a880	docs(calendar): refine attendee guidance for bots and user-search identity (#2086 ) Consolidate the user-search identity note into SKILL.md, and clarify bot handling across attendee flows: bots are virtual identities with no free/busy semantics, no meeting-room seat, and no room preference, so they must be excluded from +suggestion, +room-find, and the scheduling free/busy check. Note in create/update that bots remain valid attendees.	2026-07-28 20:34:09 +08:00
ILUO	927b37cd63	docs(task): document create data passthrough (#2080 )	2026-07-28 20:26:35 +08:00
zhangjun-bytedance	d2e22c5fca	feat: 0728 fix url (#2079 )	2026-07-28 19:05:47 +08:00
ethan-zhx	fdae560014	docs(slides): add formula inline element syntax to quick-ref (#2077 ) * docs(slides): add formula inline element syntax to quick-ref * docs(slides): add chart gradient syntax to quick-ref	2026-07-28 17:40:54 +08:00
zhengzhijiej-tech	1b173e1953	fix(sheets): recognize OFL0X local office tokens (#2063 )	2026-07-28 15:09:42 +08:00
ethan-zhx	57db1b3a8d	feat(slides):update xsd (#2067 )	2026-07-28 14:43:15 +08:00
calendar-assistant	4c1c5f5287	docs(calendar): clarify identity selection by event ownership (#2071 ) Reframe the identity section around event ownership: use `--as user` for the logged-in user's own events and `--as bot` for events the bot creates or participates in, with matching `+agenda` examples.	2026-07-28 14:05:21 +08:00
liangshuo-1	3d2c10cd0b	fix(ci): validate static workflow identity (#2015 )	2026-07-27 19:39:11 +08:00
liangshuo-1	03de81c5f3	chore: release v1.0.78 (#2061 )	2026-07-27 19:17:53 +08:00
yballul-bytedance	7abcaa7f68	feat(drive): add title+body joint search guidance and Top N pagination rules (#2059 ) * feat(drive): add title+body joint search guidance and pagination rules for Top N results - Add new blockquote explaining combined title+body search: use a single --query with both keywords instead of splitting into two searches - Add rule for Top N results: N is an output cap, not --page-size; scan up to 3 pages filtering by title and summary_highlighted, read body only for title-matched candidates, stop early at N confirmed results - Add quick-reference table row for folder-scoped title+body search - Update pagination strategy rule to cover the 3-page cap for joint search in addition to the existing 5-page limit for other scenarios * feat(drive): clarify Top N search output limit * feat(drive): clarify search filters share one call --------- Co-authored-by: yballul-bytedance <273011618+yballul-bytedance@users.noreply.github.com>	2026-07-27 17:19:48 +08:00
zhangjun-bytedance	8fb2476985	0727 fix rich text (#2062 )	2026-07-27 16:17:08 +08:00
zhanghuanxu	56c9a2afd8	fix: exempt ghost text from slides lint	2026-07-27 11:59:04 +08:00
zhanghuanxu	2029189809	fix(slides):text may over flow shape	2026-07-27 11:59:04 +08:00
zhanghuanxu	ee427979a8	fix(slides): preserve info lint severity	2026-07-27 11:59:04 +08:00
zhanghuanxu	545abcbbde	fix: refine character width estimation for lark-slides text lint Replace the uniform 0.55em half-width coefficient with per-character-type coefficients, add font-family awareness (sans/serif), bold multiplier, letter-spacing support, and fix padding-aware line wrapping. - Split half-width chars into uppercase (0.57), lowercase (0.51 sans / 0.53 serif), digits (0.58), and punctuation (0.50) - Add classify_font_family() to apply slightly wider lowercase widths for serif fonts (Georgia, Source Han Serif/思源宋体, Times, etc.) - Add 5% width multiplier for bold text; detect <strong>/<b>/<i>/<em> tags and span-level bold/italic attributes in addition to content attrs - Fix estimate_text_line_count_for_text to subtract paddingLeft/paddingRight from available width before computing wrap lines - Add resolve_letter_spacing and wire letterSpacing through estimate_text_width - Extract fontFamily/bold/italic/letterSpacing into element dict during parse	2026-07-27 11:59:04 +08:00
zhanghuanxu	4a73e83f1e	fix(slides): allow chartParsedValues roundtrip tag chartParsedValues is a server-injected roundtrip child tag under chartField, not an attribute. Move it from ROUNDTRIP_SXSD_ATTRS to a new ROUNDTRIP_SXSD_TAGS set and skip the tag (and its subtree) in the SXSD tag whitelist check.	2026-07-27 11:59:04 +08:00
zhanghuanxu	7496420fa8	fix(slides): downgrade background-decoration text overflow to info Large low-alpha text underneath other text shapes is typically a background design element; treat text_may_overflow_shape as info in that case instead of warning/error.	2026-07-27 11:59:04 +08:00
zhanghuanxu	43fabdf524	fix(slides): detect letterSpacing-driven text overflow Extract letterSpacing from content/paragraph attrs and factor it into width and line-count estimates, and stop short-circuiting the shape overflow check for autoFit shapes so that letterSpacing-heavy captions under normal-auto-fit no longer escape detection.	2026-07-27 11:59:04 +08:00
zhanghuanxu	8c46c74105	fix(slides): upgrade text overflow to error above 10px threshold Text-shape overflow was always reported as a warning, which let clearly broken pages pass the lint gate. Overflow > 10px now upgrades to error; smaller overflows stay as warning to avoid flagging near-fit cases.	2026-07-27 11:59:04 +08:00
zhanghuanxu	70777c86c3	fix(slides): restrict canvas overflow checks	2026-07-27 11:59:04 +08:00