Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyzes article structure, identifies positions requiring visual aids, generates illustrations with Type × Style two-dimension approach. Use when user asks to "illustrate article", "add images", "generate images for article", or "为文章配图".
.claude/skills/freestylefly-canghe-article-illustrator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 102% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 52% | 0% |
Analyze articles, identify illustration positions, generate images with Type × Style consistency.
| Dimension | Controls | Examples | |-----------|----------|----------| | Type | Information structure | infographic, scene, flowchart, comparison, framework, timeline | | Style | Visual aesthetics | notion, warm, minimal, blueprint, watercolor, elegant |
Combine freely: --type infographic --style blueprint
| Type | Best For | |------|----------| | infographic | Data, metrics, technical | | scene | Narratives, emotional | | flowchart | Processes, workflows | | comparison | Side-by-side, options | | framework | Models, architecture | | timeline | History, evolution |
See references/styles.md for Core Styles, full gallery, and Type × Style compatibility.
- [ ] 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: Finalize1.5 Load Preferences (EXTEND.md) ⛔ BLOCKING
bashtest -f .canghe-skills/canghe-article-illustrator/EXTEND.md && echo "project" test -f "$HOME/.canghe-skills/canghe-article-illustrator/EXTEND.md" && echo "user"
| Result | Action | |--------|--------| | Found | Read, parse, display summary | | Not found | ⛔ Run first-time-setup |
Full procedures: references/workflow.md
| Analysis | Output | |----------|--------| | Content type | Technical / Tutorial / Methodology / Narrative | | Purpose | information / visualization / imagination | | Core arguments | 2-5 main points | | Positions | Where illustrations add value |
CRITICAL: Metaphors → visualize underlying concept, NOT literal image.
Full procedures: references/workflow.md
ONE AskUserQuestion, max 4 Qs. Q1-Q3 REQUIRED.
| Q | Options | |---|---------| | Q1: Type | Recommended], infographic, scene, flowchart, comparison, framework, timeline, mixed | | Q2: Density | minimal (1-2), balanced (3-5), per-section (Recommended), rich (6+) | | Q3: Style | Recommended], minimal-flat, sci-fi, hand-drawn, editorial, scene, Other | | Q4: Language | When article language ≠ EXTEND.md setting |
Full procedures: references/workflow.md
Save outline.md with frontmatter (type, density, style, image_count) and entries:
yaml## Illustration 1 **Position**: [section/paragraph] **Purpose**: [why] **Visual Content**: [what] **Filename**: 01-infographic-concept-name.png
Full template: references/workflow.md
⛔ BLOCKING: Prompt files MUST be saved before ANY image generation.
prompts/NN-{type}-{slug}.md with YAML frontmatter--prompt without saving prompt files firstdirect/style/palette)Full procedures: references/workflow.md
Insert  after paragraphs.
Article Illustration Complete!
Article: [path] | Type: [type] | Density: [level] | Style: [style]
Images: X/N generatedillustrations/{topic-slug}/
├── source-{slug}.{ext}
├── references/ # if provided
├── outline.md
├── prompts/
└── NN-{type}-{slug}.pngSlug: 2-4 words, kebab-case. Conflict: append -YYYYMMDD-HHMMSS.
| Action | Steps | |--------|-------| | Edit | Update prompt → Regenerate → Update reference | | Add | Position → Prompt → Generate → Update outline → Insert | | Delete | Delete files → Remove reference → Update outline |
| File | Content | |------|---------| | references/workflow.md | Detailed procedures | | references/usage.md | Command syntax | | references/styles.md | Style gallery | | references/prompt-construction.md | Prompt templates | | references/config/first-time-setup.md | First-time setup |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 15,155 | 3,852 | -75% | 1 | 1 | 0% | 1,760 | 2,069 | +18% | 0 | 0 | — |
case-01 | fail→fail | 7,171 | 6,205 | -13% | 1 | 1 | 0% | 1,320 | 1,745 | +32% | 0 | 0 | — |
case-02 | fail→fail | 31,210 | 5,677 | -82% | 1 | 1 | 0% | 6,222 | 1,720 | -72% | 0 | 0 | — |
case-03 | fail→fail | 16,970 | 6,124 | -64% | 1 | 1 | 0% | 3,438 | 1,773 | -48% | 0 | 0 | — |
case-04 | pass→pass | 9,854 | 9,305 | -6% | 1 | 1 | 0% | 1,955 | 2,288 | +17% | 0 | 0 | — |
case-05 | pass→pass | 25,169 | 20,723 | -18% | 1 | 1 | 0% | 5,896 | 6,157 | +4% | 0 | 0 | — |
case-06 | pass→fail | 31,741 | 8,006 | -75% | 1 | 1 | 0% | 6,174 | 1,680 | -73% | 0 | 0 | — |
case-07 | pass→pass | 6,802 | 3,761 | -45% | 1 | 1 | 0% | 1,008 | 1,955 | +94% | 0 | 0 | — |
case-08 | fail→pass | 12,778 | 15,128 | +18% | 1 | 1 | 0% | 1,784 | 3,610 | +102% | 0 | 0 | — |
case-09 | fail→fail | 11,180 | 9,186 | -18% | 1 | 1 | 0% | 1,969 | 2,003 | +2% | 0 | 0 | — |
case-10 | fail→pass | 10,920 | 2,681 | -75% | 1 | 1 | 0% | 1,687 | 1,816 | +8% | 0 | 0 | — |
case-12 | fail→pass | 12,278 | 5,387 | -56% | 1 | 1 | 0% | 2,038 | 2,240 | +10% | 0 | 0 | — |
case-13 | pass→pass | 9,600 | 4,011 | -58% | 1 | 1 | 0% | 1,434 | 2,074 | +45% | 0 | 0 | — |
case-14 | fail→pass | 7,418 | 3,560 | -52% | 1 | 1 | 0% | 1,297 | 1,973 | +52% | 0 | 0 | — |
case-15 | fail→pass | 8,926 | 2,655 | -70% | 1 | 1 | 0% | 1,496 | 1,791 | +20% | 0 | 0 | — |
case-16 | fail→pass | 4,700 | 10,132 | +116% | 1 | 1 | 0% | 691 | 2,420 | +250% | 0 | 0 | — |
case-17 | fail→fail | 7,073 | 5,980 | -15% | 1 | 1 | 0% | 1,178 | 2,450 | +108% | 0 | 0 | — |
case-18 | fail→pass | 10,625 | 5,724 | -46% | 1 | 1 | 0% | 1,729 | 2,343 | +36% | 0 | 0 | — |
case-19 | fail→pass | 6,365 | 4,424 | -30% | 1 | 1 | 0% | 995 | 2,077 | +109% | 0 | 0 | — |
case-20 | pass→pass | 9,390 | 3,068 | -67% | 1 | 1 | 0% | 1,388 | 1,833 | +32% | 0 | 0 | — |
case-21 | fail→pass | 7,823 | 3,947 | -50% | 1 | 1 | 0% | 1,161 | 2,080 | +79% | 0 | 0 | — |
case-22 | fail→pass | 10,205 | 3,758 | -63% | 1 | 1 | 0% | 1,396 | 1,960 | +40% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +45 percentage points is the difference between those two pass rates over the 17 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.