agentleFS
JimLiuJimLiu/baoyu-skills/skills/baoyu-xhs-images/SKILL.md

baoyu-xhs-images

Skill27k starsChanged 7 months agoMITView on GitHub

Generates infographic image card series with 12 visual styles, 8 layouts, and 3 color palettes. Breaks content into 1-10 cartoon-style image cards optimized for social media engagement. Use when user mentions "小红书图片", "小红书种草", "小绿书", "微信图文", "微信贴图", "image cards", "图片卡片", baoyu-xhs-images, or wants social media infographic series.

15 other projects use a version of this, 6 of them edited; 6 of 6 edited versions we compared changed its rules; 1 new project in the last 4 weeks, against 0 the 4 before.

How people change it

6 of 6 edited versions changed its rules, and some changes recur.

Of the 6 edited versions we compared: 6 made changes of their own.

Rules people drop

  1. Codex via codex exec (codex-imagegen): no native imagegen, but the codex CLI is on PATH with an active codex login → baoyu-image-gen --provider codex-cli (preferred), or the bundled wrapper if baoyu-image-gen is unavailable. …dropped by 6 of 6
  2. ⛔ Raster only: never substitute SVG, HTML/CSS, canvas, or other code-based rendering for a raster image, even for diagram-like content. If no raster backend resolves, go to step 4.dropped by 6 of 6
  3. When the user picks a non-default style + layout combo, check references/style-layout-matrix.md and flag poor matches (✗).dropped by 6 of 6
  4. Only when the user supplies refs (--ref <files...> or pasted paths): read references/reference-images.md for intake, usage modes (direct / style / palette), prompt frontmatter, and how refs combine with the image-1 anchor chain.dropped by 6 of 6
  5. When the user asks to edit, add, delete, or fix text on a card after generation, read references/image-modification.md first. Always update the prompt file before regenerating, and never patch bitmap text with code.dropped by 6 of 6
  6. When this skill prompts the user, follow this tool-selection rule (priority order):dropped by 2 of 6
  7. Fallback: if no such tool exists, emit a numbered plain-text message and ask the user to reply with the chosen number/answer for each question.dropped by 2 of 6
  8. Batching: if the tool supports multiple questions per call, combine all applicable questions into a single call; if only single-question, ask them one at a time in priority order.dropped by 2 of 6

Rules people add

  1. A rule we can't quote: its licence doesn't allow itadded by 4 of 6
  2. A rule we can't quote: its licence doesn't allow itadded by 4 of 6
  3. A rule we can't quote: its licence doesn't allow itadded by 4 of 6
  4. If Projects/xhs-posts/{topic-slug}/ already exists:added by 2 of 6
  5. CRITICAL: If EXTEND.md not found, MUST complete first-time setup before ANY other questions or steps. Do NOT proceed to content analysis, do NOT ask about style, do NOT ask about layout — ONLY complete the preferences setup first.added by 2 of 6
  6. Check EXTEND.md existence (priority order):added by 2 of 6
  7. Language: Use user's input language or saved language preference.added by 2 of 6
  8. Use AskUserQuestion with ALL questions in ONE call. See .agents/skills/baoyu-xhs-images/references/config/first-time-setup.md for question details.added by 2 of 6

We split this file and each edited version into the rules they give an agent and match them by wording. A version that adds no rule of its own and has not changed since this file last did reads as an older text, so what it lacks is not counted as dropped. We keep no past texts of a file, so that is inferred, and an edited version that started from an older text still counts its gaps as drops.

Adopted by

Surging: 1 new project took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 1 new projectWeek of Oct 25: 0 new projectsWeek of Nov 1: 0 new projectsWeek of Nov 8: 0 new projects
Sep 2026Oct 2026Nov 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 1 in all, so this is a hint, not a trend.

Most recently taken up by

Who's talking about JimLiu/baoyu-skills

The biggest 5 of 66 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 66

Talked about on GitHub, YouTube, Bluesky and the web (66 times).

On GitHub · 50

45 more on GitHub

On YouTube · 12

7 more on YouTube

On Bluesky · 3

On the web · 1

Used with

Projects with it also use deepl, imagemagick-conversion and 8 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • zephyrwang6Skill

    deepl

    Übersetzt Texte, Dokumente und XLIFF-Dateien via DeepL API. Verwende für Übersetzungen mit /deepl translate, /deepl file oder /deepl xliff.

    Also in 4 projects that use this

  • zephyrwang6Skill

    imagemagick-conversion

    Convert and manipulate images with ImageMagick. Covers format conversion, resizing, batch processing, quality adjustment, and image transformations. Use when user mentions image conversion, resizing images, ImageMagick, magick command, batch image processing, or thumbnail generation.

    Also in 4 projects that use this

  • +8 more from zephyrwang6

The file

Show its rules and text

We split this skill into the rules it gives an agent

25 rules. 0 are common, 14 are distinctive, and 11 are very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 14

    Distinctive

    This author's own choices. Worth a look.

  • 11

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When this skill prompts the user, follow this tool-selection rule (priority order):

    line 16
  2. Distinctive:

    Fallback: if no such tool exists, emit a numbered plain-text message and ask the user to reply with the chosen number/answer for each …

    line 19
  3. Distinctive:

    Batching: if the tool supports multiple questions per call, combine all applicable questions into a single call.

    line 20
  4. Very specific:

    Codex via codex exec (codex-imagegen): no native imagegen, but the codex CLI is on PATH with an active codex login → baoyu-image-gen …

    line 32
  5. Distinctive:

    ⛔ Raster only: never substitute SVG, HTML/CSS, canvas, or other code-based rendering for a raster image, even for diagram-like content.

    line 39
  6. Very specific:

    Prompt file (hard): before invoking any backend, write each image's full, final prompt to its own file under prompts/ (NN-{type}-[slug].md) …

    line 43
  7. Distinctive:

    After every prompt file for the current generation group has been saved and verified, generate images in batches by default.

    line 49
  8. Distinctive:

    Use the chosen backend's native batch / multi-task interface if it exists.

    line 53
  9. Very specific:

    If no native batch interface exists but the runtime can issue parallel tool calls, dispatch up to generation_batch_size images at a time.

    line 54
  10. Distinctive:

    If neither native batch nor parallel tool calls are available, generate sequentially.

    line 55
  11. Distinctive:

    Never start a batch until every selected prompt file for that batch exists on disk.

    line 60
  12. Distinctive:

    Do not use subagents merely to parallelize image rendering.

    line 62
  13. Distinctive:

    Default behavior: confirm before generation.

    line 66
  14. Very specific:

    Treat explicit skill invocation, a file path, matched signals/presets, and EXTEND.md defaults as recommendation inputs only.

    line 68
  15. Distinctive:

    Do not start Step 3 until the user completes Step 2.

    line 69
  16. Distinctive:

    If confirmation is skipped explicitly, state the assumed strategy / style / layout / palette / count / backend in the next user-facing …

    line 71
  17. Very specific:

    When the user picks a non-default style + layout combo, check references/style-layout-matrix.md and flag poor matches (✗).

    line 174
  18. Very specific:

    Only when the user supplies refs (--ref <files...> or pasted paths): read references/reference-images.md for intake, usage modes (direct / …

    line 188
  19. Distinctive:

    Check these paths in order.

    line 219
  20. Very specific:

    Run the deep analysis in references/workflows/analysis-framework.md: content type, hook potential, audience, engagement signals, visual …

    line 236
  21. Very specific:

    Write everything to analysis.md.

    line 239
  22. Distinctive:

    Display summary: before asking:

    line 247
  23. Very specific:

    Path B — Customize: ask five questions (strategy/style, layout, palette, count, optional notes) with the recommendation pre-filled — blanks …

    line 266
  24. Very specific:

    Write the full prompt for every image to prompts/NN-{type}-{slug}.md in the user's preferred language (backup rule applies), then verify …

    line 282
  25. Very specific:

    When the user asks to edit, add, delete, or fix text on a card after generation, read references/image-modification.md first.

    line 339

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Image Card Series Generator

Break down complex content into eye-catching image card series with multiple style options.

User Input Tools

When this skill prompts the user, follow this tool-selection rule (priority order):

  1. Prefer built-in user-input tools exposed by the current agent runtime — e.g., AskUserQuestion, request_user_input, clarify, ask_user, or any equivalent.
  2. Fallback: if no such tool exists, emit a numbered plain-text message and ask the user to reply with the chosen number/answer for each question.
  3. Batching: if the tool supports multiple questions per call, combine all applicable questions into a single call; if only single-question, ask them one at a time in priority order.

Concrete AskUserQuestion references below are examples — substitute the local equivalent in other runtimes.

Image Generation Tools

Resolve the image backend in this order:

  1. User names a backend in the current message → use it.
  2. preferred_image_backend in EXTEND.md names a backend available now → use it. ask means ask every run.
  3. Auto-select (preference auto, unset, or the pinned backend is unavailable), first match wins:
    • Codex imagegen skill is in your available skills → you are in Codex and MUST use it: Skill with skill: "imagegen", passing the prompt file's content plus output path and aspect ratio. It outranks non-native skills such as baoyu-image-gen.
    • Codex via codex exec (codex-imagegen): no native imagegen, but the codex CLI is on PATH with an active codex login → baoyu-image-gen --provider codex-cli (preferred), or the bundled wrapper if baoyu-image-gen is unavailable. Load references/codex-imagegen.md only for this branch.
    • Cursor GenerateImage → use it (outranks non-native skills). It has no aspect-ratio parameter, so state the aspect ratio / dimensions in the description prompt; it saves to a tool-managed location, so copy the file to the expected output path (e.g., outputs/.../NN-xxx.png). Reference images go in reference_image_paths.
    • Other runtime-native image tool (e.g., Hermes image_generate) → use it the same way.
    • Exactly one non-native backend installed (e.g., baoyu-image-gen) → use it.
    • Several non-native backends → ask the user once, batched with other initial questions.
  4. None available → tell the user and ask how to proceed.

⛔ Raster only: never substitute SVG, HTML/CSS, canvas, or other code-based rendering for a raster image, even for diagram-like content. If no raster backend resolves, go to step 4.

⛔ No bitmap text patching: never use ImageMagick, Pillow, Canvas, SVG, HTML/CSS, OCR scripts, or any other overlay to cover, rewrite, erase, stroke, or replace titles, body copy, tags, or any other text inside an already generated image card. Regenerate from a corrected prompt, switch to a layout with less on-card text, or ask the user which imperfect candidate to keep.

Prompt file (hard): before invoking any backend, write each image's full, final prompt to its own file under prompts/ (NN-{type}-[slug].md) and pass that file (or its content) to the backend. It is the reproducibility record and lets you switch backends without rewriting prompts.

Change the pinned backend via ## Changing Preferences below. Tool names above are examples; use the local equivalent in other runtimes.

Batch Generation Policy

After every prompt file for the current generation group has been saved and verified, generate images in batches by default.

Priority order:

  1. Use the chosen backend's native batch / multi-task interface if it exists. Each task must keep its own prompt file, output path, aspect ratio, session ID, and direct reference images.
  2. If no native batch interface exists but the runtime can issue parallel tool calls, dispatch up to generation_batch_size images at a time. Default: 4. An explicit user request in the current message, such as --batch-size 4 or "并行 4 张一起生成", overrides EXTEND.md.
  3. If neither native batch nor parallel tool calls are available, generate sequentially.

Rules:

  • Honor the image-1 anchor chain: generate image 1 first, then batch images 2+ using image 1 as the reference.
  • Never start a batch until every selected prompt file for that batch exists on disk.
  • Retry failed items once without regenerating successful items.
  • Do not use subagents merely to parallelize image rendering. Use subagents only for separate prompt iteration or creative exploration.

Confirmation Policy

Default behavior: confirm before generation.

  • Treat explicit skill invocation, a file path, matched signals/presets, and EXTEND.md defaults as recommendation inputs only. None of them authorizes skipping confirmation.
  • Do not start Step 3 until the user completes Step 2.
  • Skip confirmation only when the current request explicitly says to do so, for example: --yes, "直接生成", "不用确认", "跳过确认", "按默认出图", or equivalent wording.
  • If confirmation is skipped explicitly, state the assumed strategy / style / layout / palette / count / backend in the next user-facing update before generating.

Language

Respond in the user's language across questions, progress, errors, and completion summary. Keep technical tokens (style names, file paths, code) in English.

Options

OptionDescription
--style <name>Visual style (see Styles below)
--layout <name>Information layout (see Layouts below)
--palette <name>Color override: macaron / warm / neon
--preset <name>Style + layout + optional palette shorthand (see references/style-presets.md)
--ref <files...>Reference images applied to image 1 as the series anchor
--batch-size <n>Temporary generation batch size for this run. Default: generation_batch_size from EXTEND.md, otherwise 4. Clamp to 1-8.
--yesNon-interactive: skip all confirmations, use EXTEND.md or built-in defaults, auto-confirm recommended plan (Path A)

Dimensions

Three independent knobs combine freely:

DimensionControlsOptions
StyleVisual aesthetics (lines, decorations, rendering)12 styles (see Styles below)
LayoutInformation structure (density, arrangement)8 layouts (see Layouts below)
Palette (optional)Color override, replaces the style's default colorsmacaron / warm / neon (see Palettes below)

Example: --style notion --layout dense makes an intellectual knowledge card; add --palette macaron to soften the colors without changing notion's rendering rules. A --preset is a shorthand for style + layout (+ optional palette).

Palette behavior: no --palette → style's built-in colors; --palette <name> → overrides colors only, rendering rules unchanged. Some styles declare a default_palette (e.g., sketch-notes defaults to macaron).

Styles (12)

StyleDescription
cute (Default)Sweet, adorable, girly aesthetic
freshClean, refreshing, natural
warmCozy, friendly, approachable
boldHigh impact, attention-grabbing
minimalUltra-clean, sophisticated
retroVintage, nostalgic, trendy
popVibrant, energetic, eye-catching
notionMinimalist hand-drawn line art, intellectual
chalkboardColorful chalk on black board, educational
study-notesRealistic handwritten photo style, blue pen + red annotations + yellow highlighter
screen-printBold poster art, halftone textures, limited colors, symbolic storytelling
sketch-notesHand-drawn educational infographic, macaron pastels on warm cream, wobble lines

Per-style specifications: references/presets/<style>.md.

Layouts (8)

LayoutDescription
sparse (Default)1-2 points, maximum impact
balanced3-4 points, standard
dense5-8 points, knowledge-card style
listEnumeration / ranking (4-7 items)
comparisonSide-by-side contrast
flowProcess / timeline (3-6 steps)
mindmapCenter-radial (4-8 branches)
quadrantFour-quadrant / circular sections

Layout specs: references/elements/canvas.md.

Palettes (optional override)

Replaces the style's colors while keeping rendering rules (line treatment, textures) intact.

PaletteBackgroundZone ColorsAccentFeel
macaronWarm cream #F5F0E8Blue #A8D8EA, Lavender #D5C6E0, Mint #B5E5CF, Peach #F8D5C4Coral #E8655ASoft, educational
warmSoft peach #FFECD2Orange #ED8936, Terracotta #C05621, Golden #F6AD55, Rose #D4A09ASienna #A0522DEarth tones, cozy
neonDark purple #1A1025Cyan #00F5FF, Magenta #FF00FF, Green #39FF14, Pink #FF6EC7Yellow #FFFF00High-energy, futuristic

Palette specs: references/palettes/<palette>.md.

Presets (style + layout shortcuts)

26 presets grouped by scenario (knowledge, lifestyle, impact, trend, poster). Read references/style-presets.md when the user passes --preset or when recommending a preset in Step 2.

Auto-Selection

Match content signals to the best combo. First row whose keywords appear wins; fall back to cute-share if nothing matches.

Signals in sourceStyleLayoutRecommended preset
beauty, fashion, cute, girl, pinkcutesparse/balancedcute-share, girly
health, nature, fresh, organicfreshbalanced/flowproduct-review, nature-flow
life, story, emotion, warmwarmbalancedcozy-story
warning, important, must, criticalboldlist/comparisonwarning, versus
professional, business, elegantminimalsparse/balancedclean-quote, pro-summary
classic, vintage, traditionalretrobalancedthrowback, retro-ranking
fun, exciting, wow, amazingpopsparse/listhype, pop-facts
knowledge, concept, productivity, SaaSnotiondense/listknowledge-card, checklist
education, tutorial, learning, classroomchalkboardbalanced/densetutorial, classroom
notes, handwritten, study guide, realisticstudy-notesdense/list/mindmapstudy-guide
movie, poster, opinion, editorial, cinematicscreen-printsparse/comparisonposter, editorial, cinematic
hand-drawn, infographic, workflow, 手绘,图解sketch-notesflow/balanced/densehand-drawn-edu, sketch-card, sketch-summary

Style × Layout Matrix

When the user picks a non-default style + layout combo, check references/style-layout-matrix.md and flag poor matches (✗).

Outline Strategies

Three differentiated approaches — each produces a structurally different outline. The workflow recommends one; Path C generates all three and lets the user choose.

StrategyConceptBest forStructure
A — Story-DrivenPersonal experience as the thread, emotional resonance firstReviews, personal shares, transformationHook → Problem → Discovery → Experience → Conclusion
B — Information-DenseValue-first, efficient information deliveryTutorials, comparisons, checklistsCore conclusion → Info card → Pros/Cons → Recommendation
C — Visual-FirstVisual impact as core, minimal textHigh-aesthetic products, lifestyle, mood contentHero image → Detail shots → Lifestyle scene → CTA

Reference Images

Only when the user supplies refs (--ref <files...> or pasted paths): read references/reference-images.md for intake, usage modes (direct / style / palette), prompt frontmatter, and how refs combine with the image-1 anchor chain.

File Layout

image-cards/{topic-slug}/
├── source-{slug}.{ext}
├── analysis.md
├── outline-strategy-{a,b,c}.md    # Path C only
├── outline.md
├── prompts/NN-{type}-{slug}.md
├── NN-{type}-{slug}.png
└── refs/                          # only if --ref used

Slug: 2-4 words, kebab-case. "AI 工具推荐" → ai-tools-recommend. On collision, append -YYYYMMDD-HHMMSS.

Backup rule (applies throughout): before overwriting any file — source, outline, prompt, image — rename the existing one to <name>-backup-YYYYMMDD-HHMMSS.<ext>. This protects user edits.

Workflow

- [ ] Step 0: Load EXTEND.md ⛔ BLOCKING (interactive only)
- [ ] Step 1: Analyze content → analysis.md
- [ ] Step 2: Smart Confirm ⚠️ REQUIRED (Path A / B / C)
- [ ] Step 3: Generate images
- [ ] Step 4: Completion report

Step 0: Load EXTEND.md ⛔ BLOCKING

Check these paths in order; first hit wins:

PathScope
.baoyu-skills/baoyu-xhs-images/EXTEND.mdProject
${XDG_CONFIG_HOME:-$HOME/.config}/baoyu-skills/baoyu-xhs-images/EXTEND.mdXDG
$HOME/.baoyu-skills/baoyu-xhs-images/EXTEND.mdUser home
  • Found → read, parse, print a summary (style / layout / watermark / language), continue.
  • Not found + interactive → run first-time setup (see references/config/first-time-setup.md) and save before anything else. Do NOT analyze content or ask style questions until preferences exist — this keeps first-run behavior predictable.
  • Not found + --yes → skip setup, use built-in defaults (no watermark, style/layout auto-selected, language from content). Do not prompt, do not create EXTEND.md.

EXTEND.md keys: watermark, preferred style/layout, custom style definitions, language preference, preferred image backend, generation batch size. Schema: references/config/preferences-schema.md.

Step 1: Analyze Content → analysis.md

  1. Save the source (backup rule applies if source.md exists).
  2. Run the deep analysis in references/workflows/analysis-framework.md: content type, hook potential, audience, engagement signals, visual opportunity map, swipe flow.
  3. Detect source language, pick recommended image count (2-10).
  4. Auto-recommend strategy + style + layout + palette using the Auto-Selection table above.
  5. Write everything to analysis.md.

Step 2: Smart Confirm ⚠️ REQUIRED

Hard gate: this step is mandatory per the Confirmation Policy — Step 3 cannot start until the user confirms here (or explicitly opts out with --yes / equivalent wording in the current request).

Goal: present the auto-recommended plan and let the user confirm or adjust. Skip this step entirely under --yes — proceed with Path A using the analysis and any CLI overrides.

Display summary before asking:

📋 内容分析
  主题:[topic] | 类型:[content_type]
  要点:[key points]
  受众:[audience]

🎨 推荐方案(自动匹配)
  策略:[A/B/C] [name]([reason])
  风格:[style] · 布局:[layout] · 配色:[palette or 默认] · 预设:[preset]
  图片:[N]张(封面+[N-2]内容+结尾)
  元素:[background] / [decorations] / [emphasis]

Then ask one question — three paths. Verbatim option copy: references/confirmation.md.

Path A — Quick confirm (trust auto-recommendation): generate a single outline using the recommended strategy + style → save to outline.md → Step 3.

Path B — Customize: ask five questions (strategy/style, layout, palette, count, optional notes) with the recommendation pre-filled — blanks keep the recommendation. Generate one outline with the user's choices → outline.md → Step 3. See references/confirmation.md.

Path C — Detailed mode: two sub-confirmations.

  • Step 2a — Content understanding: ask selling points (multi-select), audience, style preference (authentic / professional / aesthetic / auto), optional context. Update analysis.md.
  • Step 2b — Three outline variants: generate outline-strategy-a.md, outline-strategy-b.md, outline-strategy-c.md. Each MUST have a different structure AND a different recommended style — include style_reason in the frontmatter. Page-count heuristic: A ~4-6, B ~3-5, C ~3-4. Template: references/workflows/outline-template.md; frontmatter example in references/confirmation.md.
  • Step 2c — Selection: ask three questions (outline A/B/C/Combined, style, visual elements). Save selected/merged outline to outline.md → Step 3.

Step 3: Generate Images

With confirmed outline + style + layout + palette:

Visual consistency — image-1 anchor chain: character / mascot / color rendering drifts between calls unless you anchor them. Generate image 1 (cover) first WITHOUT --ref, then pass image 1 as --ref to every subsequent image. This is the single most important consistency trick for this skill — don't skip it even if the backend also supports a session ID.

Generation flow:

  1. Write the full prompt for every image to prompts/NN-{type}-{slug}.md in the user's preferred language (backup rule applies), then verify all selected prompt files exist.
  2. Generate image 1 first without --ref; backup rule applies to the PNG file. This establishes the anchor.
  3. Build a task list for images 2+ using image 1 as --ref <path-to-image-01.png>.
  4. Dispatch images 2+ in batches per the ## Batch Generation Policy: backend native batch first, runtime parallel tool calls second, sequential only as fallback.
  5. Report progress after each completed image. On failure, retry only the failed item once from the same saved prompt file.

Watermark (if enabled in EXTEND.md): append to the generation prompt:

Include a subtle watermark "[content]" positioned at [position].
The watermark should be legible but not distracting.

See references/config/watermark-guide.md.

Backend selection: per the Image Generation Tools rule at the top — use whatever is available, ask once if multiple, before any generation. Under --yes, use the EXTEND.md preference and fall back to the first available backend. Prompt files MUST exist before invoking any backend.

codex-imagegen invocation: when the rule resolves to codex-imagegen, see references/codex-imagegen.md for the invocation contract (preferred baoyu-image-gen --provider codex-cli path, runtime wrapper discovery, parameter notes, stdout schema, batch semantics — n=1 per call so card batches must dispatch one wrapper call per card; the wrapper does NOT accept --sessionId, so chain consistency must come from --ref per Step 3 above).

Session ID (if the backend supports --sessionId): use cards-{topic-slug}-{timestamp} for every image; combined with the ref chain this gives maximum consistency.

Step 4: Completion Report

Image Card Series Complete!

Topic: [topic]
Mode: [Quick / Custom / Detailed]
Strategy: [A/B/C/Combined]
Style: [name]
Palette: [name or "default"]
Layout: [name or "varies"]
Location: [directory]
Images: N total

✓ analysis.md
✓ outline.md
✓ outline-strategy-a/b/c.md (detailed mode only)

- 01-cover-[slug].png ✓ Cover (sparse)
- 02-content-[slug].png ✓ Content (balanced)
- ...
- NN-ending-[slug].png ✓ Ending (sparse)

Content Breakdown Principles

PositionPurposeTypical layout
Cover (image 1)Hook + visual impactsparse
Content (middle)Core value per imagebalanced / dense / list / comparison / flow
Ending (last)CTA / summarysparse or balanced

For the style × layout compatibility matrix, see references/style-layout-matrix.md.

Image Modification

When the user asks to edit, add, delete, or fix text on a card after generation, read references/image-modification.md first. Always update the prompt file before regenerating, and never patch bitmap text with code.

References

FileContent
references/confirmation.mdVerbatim AskUserQuestion copy for every confirmation path
references/style-presets.mdFull preset shortcut definitions, presets grouped by scenario
references/style-layout-matrix.mdStyle × layout compatibility scores
references/reference-images.mdUser-supplied reference image intake and usage
references/image-modification.mdEdit / add / delete cards, text correction policy
references/presets/<style>.mdPer-style element definitions
references/palettes/<name>.mdPer-palette color definitions
references/elements/canvas.mdAspect ratios, safe zones, grid layouts
references/elements/image-effects.mdCutout, stroke, filters
references/elements/typography.mdDecorated text, tags, text direction
references/elements/decorations.mdEmphasis marks, backgrounds, doodles, frames
references/workflows/analysis-framework.mdContent analysis framework
references/workflows/outline-template.mdOutline template with layout guide
references/workflows/prompt-assembly.mdPrompt assembly guide
references/config/preferences-schema.mdEXTEND.md schema
references/config/first-time-setup.mdFirst-time setup flow
references/config/watermark-guide.mdWatermark configuration

Notes

  • Auto-retry once on generation failure before reporting an error.
  • For sensitive public figures, use stylized cartoon alternatives.
  • Smart Confirm (Step 2) is required; Detailed mode adds a second confirmation (2a + 2c).

Changing Preferences

EXTEND.md lives at the first matching path listed in Step 0. Three ways to change it:

  • Edit directly — open EXTEND.md and change fields. Full schema: references/config/preferences-schema.md.
  • Reconfigure interactively — delete EXTEND.md (or ask "reconfigure baoyu-xhs-images preferences" / "重新配置"). The next run re-triggers first-time setup.
  • Common one-line edits:
    • preferred_image_backend: auto — default; runtime-native tool wins, falls back to the only installed backend, asks only if multiple non-native are present.
    • preferred_image_backend: codex-imagegen — pin to Codex's built-in.
    • preferred_image_backend: baoyu-image-gen — pin to the baoyu-image-gen skill.
    • preferred_image_backend: ask — confirm backend every run.
    • generation_batch_size: 4 — default number of images to render concurrently when the backend/runtime supports batch or parallel generation.
    • preferred_style: notion, preferred_layout: dense, preferred_palette: macaron, language: zh.
    • watermark.enabled: true + watermark.content: "@handle" — add a watermark.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.