Article Illustrator

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 labels, captions, or any other text inside an already generated illustration. Regenerate from a corrected prompt, redraw with less or no on-image 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 run has been saved and verified, generate images in batches by default:

1. Use the chosen backend's native batch / multi-task interface if it exists. Each task keeps its own prompt file, output path, aspect ratio, and direct reference images.
2. Otherwise, if the runtime can issue parallel tool calls, dispatch up to generation_batch_size images at a time (default 4). An explicit request in the current message (--batch-size 4, "并行4张一起生成") overrides EXTEND.md.
3. Otherwise generate sequentially.

• Never start a batch until all its prompt files exist on disk.
• Retry failed items once without regenerating successful ones.
• Don't use subagents just to parallelize rendering; only for separate per-image prompt iteration or creative exploration.

Confirmation Policy

Default: confirm before generation.

• Explicit skill invocation, a file path, matched signals/presets, and EXTEND.md defaults are recommendation inputs only — none authorizes skipping confirmation.
• Do not start Step 4 or later until the user completes Step 3.
• Skip confirmation only when the current request explicitly says so ("直接生成", "不用确认", "跳过确认", "按默认出图", or equivalent). Then state the assumed type / density / style / palette / language / backend in the next user-facing update before generating.

Reference Images

Users may supply refs via --ref <files...>, file paths, or pasted images; they guide style, palette, composition, or subject. Step 1.0 in references/workflow.md saves them to references/NN-ref-{slug}.{ext}; Step 5.3 processes per-illustration usage direct | style | palette. When the backend supports batch input, propagate direct entries from each prompt file's references: frontmatter into its batch task (e.g. baoyu-image-gen accepts ref per task).

Three Dimensions

Combine freely (--type infographic --style vector-illustration --palette macaron) or use --preset edu-visual for type + style + palette in one flag (references/style-presets.md). Command syntax and options: references/usage.md.

Workflow

- [ ] Step 1: Pre-check (EXTEND.md, references, config)
- [ ] Step 2: Analyze content
- [ ] Step 3: Confirm settings (AskUserQuestion)
- [ ] Step 4: Generate outline
- [ ] Step 5: Generate images
- [ ] Step 6: Finalize

Detailed procedures for every step: references/workflow.md.

Step 1: Pre-check

1.5 Load Preferences (EXTEND.md) ⛔ BLOCKING — first match wins:

Found → read, parse, display summary. Not found → ⛔ run first-time-setup before anything else.

Also: save reference images (1.0), determine input type and output dir (1.1–1.4). See workflow.md.

Step 2: Analyze

Determine content type (Technical / Tutorial / Methodology / Narrative), purpose (information / visualization / imagination), 2-5 core arguments, and positions where illustrations add value. Metaphors → visualize the underlying concept, never the literal image. See workflow.md.

Step 3: Confirm Settings ⚠️

Hard gate per the Confirmation Policy. ONE AskUserQuestion, max 4 Qs. Q1-Q2 required; Q3 required unless a preset was chosen.

See workflow.md.

Step 4: Generate Outline

Save outline.md with frontmatter (type, density, style, palette, image_count) and one entry per illustration (Position, Purpose, Visual Content, Filename 01-infographic-concept-name.png). Template: workflow.md.

Step 5: Generate Images

⛔ Prompt files MUST be saved before ANY image generation, whatever the backend.

1. Per illustration, write prompts/NN-{type}-{slug}.md (YAML frontmatter) per references/prompt-construction.md: type-specific template with ZONES / LABELS / COLORS / STYLE / ASPECT; LABELS use article-specific data (actual numbers, terms, metrics, quotes). Never pass ad-hoc inline prompts to --prompt without saving prompt files first.
2. Select the backend once per session via ## Image Generation Tools. For codex-imagegen, see references/codex-imagegen.md.
3. Process references (direct/style/palette) per prompt frontmatter; apply watermark if EXTEND.md enables it.
4. Generate from saved prompt files per ## Batch Generation Policy; retry once on failure.

See workflow.md.

Step 6: Finalize

Insert ![description]({relative-path}/NN-{type}-{slug}.png) after the corresponding paragraph (path relative to the article file, per Output Directory), then print the completion summary (workflow.md).

Output Directory

Set by default_output_dir in EXTEND.md:

{output-dir}/
├── outline.md
├── prompts/
│   └── NN-{type}-{slug}.md
└── NN-{type}-{slug}.png

Pasted content (no file path) always uses illustrations/{topic-slug}/ with source-{slug}.{ext} saved alongside. Slug: 2-4 words, kebab-case. Conflict: append -YYYYMMDD-HHMMSS.

Modification

Text correction: never patch misspelled, garbled, or weak rendered text in the bitmap with code. Write a new prompt file and a new output path so the flawed candidate is kept for comparison. Post-processing is limited to crop, resize, compression, or format conversion that doesn't alter text or composition.

Changing Preferences

EXTEND.md lives at the first matching path in Step 1.5. Edit it directly (schema: references/config/preferences-schema.md), or delete it / ask "reconfigure baoyu-article-illustrator preferences" / "重新配置" to re-run first-time setup. Common edits:

• preferred_image_backend: auto (default; runtime-native wins, else the only installed backend, ask if multiple non-native) | codex-imagegen | baoyu-image-gen | ask (confirm every run)
• generation_batch_size: 4 — concurrent renders when parallel calls are supported
• preferred_type: infographic, preferred_style: notion, preferred_palette: macaron, language: zh
• default_output_dir: imgs-subdir