baoyu-slide-deck skill

Slide Deck Generator

Transform content into professional slide deck images. The deck is designed for reading and sharing (self-explanatory slides, logical scroll flow, social-media-friendly) rather than live presentation — that assumption drives every layout and density decision below.

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, requestuserinput, 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

When this skill needs to render an image, resolve the backend in this order:

1. Current-request override — if the user names a specific backend in the current message, use it.
2. Saved preference — if EXTEND.md sets preferredimagebackend to a backend available right now, use it.
3. Auto-select (when the preference is auto, unset, or the pinned backend isn't available):

• Codex (imagegen) — first, inspect your available-skills / tool inventory. If a skill named imagegen is listed, you are running inside Codex and MUST use it: invoke via the Skill tool with skill: "imagegen", passing the saved prompt file's content (plus output path and aspect ratio per Codex imagegen's own args). Codex imagegen is the official raster backend in that runtime and outranks any non-native skill (e.g., baoyu-image-gen) unless the user has explicitly pinned a different preferredimagebackend.
• Codex via codex exec (codex-imagegen) — if the current runtime exposes no native imagegen skill but the codex CLI is on PATH with an active codex login, route through baoyu-image-gen --provider codex-cli (preferred), or — if baoyu-image-gen is unavailable — invoke the bundled wrapper directly. Details, parameters, and the runtime-discovery procedure live in references/codex-imagegen.md — load that file only when this branch is selected.
• Cursor (GenerateImage) — if the runtime exposes a native GenerateImage tool, you are running inside Cursor and it outranks any non-native skill the same way Codex imagegen does. Two hard caveats: (a) it has no aspect-ratio parameter — state the target aspect ratio / dimensions explicitly in the prompt text passed as description; (b) it does not accept an output directory — it saves to a tool-managed location, so after generation copy/move the file to the skill's expected output path (e.g., outputs/.../NN-xxx.png). Reference images go in referenceimagepaths.
• Other runtime-native tools — if the runtime exposes a different native image tool (e.g., Hermes image_generate), use it the same way.
• Otherwise, if exactly one non-native backend is installed (e.g., baoyu-image-gen), use it.
• Otherwise (multiple non-native backends with no runtime-native tool), ask the user once — batch with any other initial questions.

1. If none are available, tell the user and ask how to proceed.

⛔ Never substitute SVG, HTML, canvas, or other code-based rendering for raster image generation. Codex imagegen's own description says it should be used "when the output should be a bitmap asset rather than repo-native code or vector." If you cannot resolve a raster backend via step 3, fall through to step 4 and ask the user — do not silently emit SVG, write inline markup, or produce HTML/CSS art as a substitute. This applies even if the article/section seems "diagram-like": the consumer skill calling this rule has already decided that a raster image is what it needs.

⛔ Never repair rendered text by painting over a generated bitmap. Do not use ImageMagick, Pillow, Canvas, SVG, HTML/CSS, OCR scripts, or any other programmatic overlay to cover, rewrite, erase, stroke, or replace slide titles, bullets, or any other text inside an already generated slide image. If text is wrong or unclear, regenerate from a corrected prompt, simplify the slide's on-image text, or ask the user which imperfect candidate to keep.

Setting preferredimagebackend: ask forces the step-3 prompt every run regardless of available backends. Users change the pinned backend via the ## Changing Preferences section below.

Prompt file requirement (hard): write each image's full, final prompt to a standalone file under prompts/ (naming: NN-slide-[slug].md) BEFORE invoking any backend. The file is the reproducibility record and lets you switch backends without regenerating prompts.

Concrete tool names (imagegen, GenerateImage, image_generate, baoyu-image-gen) above are examples — substitute the local equivalents under the same rule.

Batch Generation Policy

After every prompt file for the current generation group has been saved and verified, generate slide 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 generationbatchsize slide 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:

• Never start the first batch until all selected slide prompt files exist 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.
• Merge PPTX/PDF only after all selected slide images are generated.

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 or later until the user completes Step 2.
• Skip confirmation only when the current request explicitly says to do so, for example: "直接生成", "不用确认", "跳过确认", "按默认出幻灯片", or equivalent wording.
• If confirmation is skipped explicitly, state the assumed style / audience / slide-count / language / backend in the next user-facing update before generating.

Language

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

Script Directory

{baseDir} = this SKILL.md's directory. Resolve ${BUN_X}: prefer bun; else npx -y bun; else suggest brew install oven-sh/bun/bun.

Options

Style System

17 presets covering technical / educational / lifestyle / editorial use cases. Every preset is a combination of four dimensions (texture / mood / typography / density). If the user picks "Custom dimensions" in Round 1, Round 2 of the confirmation asks one question per dimension — options and verbatim copy live in references/confirmation.md.

Presets (17)

Per-preset specs: references/styles/.md. Preset → dimension mapping: references/dimensions/presets.md.

Dimensions (when "Custom dimensions" picked)

Full per-dimension specs: references/dimensions/*.md.

Auto-Selection

Match content signals to a preset. Pick the first row whose signal keywords appear in the source; fall back to blueprint if nothing matches.

Slide Count Heuristic

Reference Images

Users may supply reference images to guide style, palette, layout, or subject.

Intake: Accept via --ref or when the user provides file paths / pastes images in conversation.

• File path → copy to {slide-deck-dir}/refs/NN-ref-{slug}.{ext}
• Pasted image with no path → ask for the path, or extract style traits verbally as a text fallback

Usage modes (per reference):

Record refs in each slide's prompt frontmatter:

references:
  - ref_id: 01
    filename: 01-ref-brand.png
    usage: direct

At generation time, verify files exist. If usage: direct and the backend accepts refs (e.g., baoyu-image-gen --ref), pass the file on every slide. Otherwise embed extracted style/palette traits in the prompt text.

File Layout

slide-deck/{topic-slug}/
├── source-{slug}.{ext}
├── outline.md
├── prompts/NN-slide-{slug}.md
├── NN-slide-{slug}.png
├── {topic-slug}.pptx
└── {topic-slug}.pdf

Slug: 2-4 words, kebab-case, extracted from topic. "Introduction to Machine Learning" → intro-machine-learning.

Backup rule (applies across steps): if a file about to be written already exists, rename it to -backup-YYYYMMDD-HHMMSS. before writing the new one. This protects user edits and enables rollback.

Workflow

Copy this checklist and check off items as you complete them:

- [ ] Step 1: Setup & analyze
- [ ] Step 2: Confirmation ⚠️ REQUIRED (Round 1; Round 2 only if "Custom dimensions")
- [ ] Step 3: Generate outline
- [ ] Step 4: Review outline (conditional)
- [ ] Step 5: Generate prompts
- [ ] Step 6: Review prompts (conditional)
- [ ] Step 7: Generate images
- [ ] Step 8: Merge to PPTX/PDF
- [ ] Step 9: Output summary

Step 1: Setup & Analyze

1.1 Load EXTEND.md — check these paths in order; first hit wins:

If found, read, parse, and print a summary (style / audience / language / review / generation batch size). If not, proceed with defaults — first-time setup is not blocking for this skill. Schema: references/config/preferences-schema.md.

1.2 Analyze content — follow references/analysis-framework.md: classify content, detect language, note signals for style selection, estimate slide count from length (see the Slide Count Heuristic in Style System above), generate topic slug. Save source as source.md (honor backup rule if one exists).

1.3 Check existing output ⚠️ REQUIRED before Step 2. If slide-deck/{topic-slug}/ exists, ask how to proceed — four options (regenerate outline / regenerate images / backup and regenerate / exit), verbatim copy in references/confirmation.md.

Save findings to analysis.md: topic, audience, signals, recommended style and slide count, language detection.