infographic
Skill18 starsChanged 4 months agoMITView on GitHub
Generate hand-drawn educational infographic images from any topic using Google Gemini image generation. Use this skill whenever the user says /infographic, 'create an infographic', 'visualize this', 'make a diagram', 'explain this visually', 'create visuals for', 'I need graphics for', 'illustrate this concept', 'drill into this', 'go deeper on', or wants to turn a topic, system, or process into educational visual graphics. Always trigger when the user wants something explained visually — even if they don't say 'infographic'.
- Reads credentials
- Installs packages
14 other projects use a version of this.
Adopted by
Most recently taken up by
- berthelol/skillstheir own version, first seen Nov 15, 2026
- CatCatUncle/openworkbuddytheir own version, first seen Nov 5, 2026
- pahlevikun/levistacktheir own version, first seen Nov 1, 2026
- arnaugonzalez/infographic-post-skilltheir own version, first seen Nov 1, 2026
- magnus919/agent-skillstheir own version, first seen Nov 1, 2026
- ArnavPuri/designskillstheir own version, first seen Nov 1, 2026
Other versions14 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 14
- ZJU-REAL/Easel★ 4.7ksame name
- claude-office-skills/skills★ 433same name
- CatCatUncle/openworkbuddy★ 413same name
- magnus919/agent-skills★ 177same name
- jincai/openclaw-skills★ 18same name
- ArnavPuri/designskills★ 17same name
- costiash/eliviz★ 7same name
- remybroun/infographic★ 4same name
- amigoscode/skills★ 1same name
- arnaugonzalez/infographic-post-skill★ 1same name
4 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository52 other files it gives its agents.
- ad-studioSkill
- brainstormingSkill
- build-siteSkill
- carousel-creatorSkill
- cold-outreachSkill
- content-engineSkill
- content-promptSkill
- daily-digestSkill
- daily-post-ideasSkill
- defuddleSkill
- digital-guide-builderSkill
- dissectSkill
- dm-writerSkill
- edu-siteSkill
- faceless-videoSkill
- first-customer-closerSkill
- frontend-designSkill
- higgsfield-generateSkill
- hooksSkill
- ig-channel-remixSkill
- launch-planSkill
- lead-magnetSkill
- linkedin-post-batchSkill
- linkedin-postSkill
- meeting-notesSkill
- money-model-auditSkill
- objection-handlerSkill
- offer-builderSkill
- one-lead-magnetSkill
- outlinesSkill
- personalizeSkill
- positioning-filterSkill
- pricing-calculatorSkill
- productize-serviceSkill
- proposal-generatorSkill
- saas-ad-studioSkill
- sanitize-projectSkill
- shortifySkill
- substackSkill
- ui-ux-pro-maxSkill
- viral-format-minerSkill
- viral-remixSkill
- warm-outreachSkill
- web-design-guidelinesSkill
- yapSkill
- youtube-clipping-agentSkill
- yt-descriptionSkill
- yt-introSkill
- yt-searchSkill
- yt-thumbnailSkill
- yt-titlesSkill
- skill-nameSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 7 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
After showing the Level 1 set, always offer:
- Distinctive: line 35
Keep the same character design across all Level 2 diagrams for consistency (lock the character description in every prompt)
- Distinctive: line 111
Run all images in parallel background tasks.
- Very specific: line 121
Run from the skill directory so the relative scripts/ path resolves (or use the absolute path to generate_image.py).
- Very specific: line 123
Default: 16:9 at 2K
- Distinctive: line 140
After Level 1: offer to drill deeper on any diagram.
- Distinctive: line 141
After Level 2: offer to drill into another component or regenerate any image.
- Distinctive: line 171
Arrows not labeled: Add each arrow label explicitly in the prompt
- Distinctive: line 172
Character looks different across images: Use the exact same robot character description verbatim in every prompt
- Very specific: line 177
Parallel generation failures: Don't run multiple scripts via shell & loops.
- Very specific: line 178
Wrong model: Only use gemini-3.1-flash-image-preview.
“Common” is measured: kept by at least 6 in 10 of the 11 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.
Files
Infographic Creator
Requirements
Image generation runs through Google Gemini via scripts/generate_image.py, which needs a Google AI (Gemini) API key. The script reads it from the environment — set GOOGLE_API_KEY (or GEMINI_API_KEY) before running (e.g. export GOOGLE_API_KEY=..., or add it to your shell profile / Claude Code settings env). Get a free key at https://aistudio.google.com/apikey. If the key isn't set, the script errors and I'll tell you exactly what to add. The Python deps install with pip install google-genai Pillow.
Turn any concept into a two-level visual explainer series:
- Level 1 — Overview: 3–5 broad diagrams that tell the full story of the topic
- Level 2 — Drill-down: For any Level 1 diagram, generate one deeper diagram per component — each with internal callout annotations showing what's happening inside
Style target: Dark charcoal background, rough sketch ink lines, white and warm accent colors, handwritten text throughout. Warm, approachable, human — like a smart teacher drew it on a dark sketchpad.
Two-Level Structure
Level 1: Overview Run
The default. Generate 3–5 diagrams that cover the full topic at a high level. Each diagram should be understandable on its own but also work as part of a set.
After showing the Level 1 set, always offer:
"Want to drill deeper into any of these? I can generate a Level 2 set for [diagram title] — one diagram per component with internal callouts."
Level 2: Drill-Down Run
Triggered when the user says "drill into [X]", "go deeper on [X]", or picks a specific Level 1 diagram to expand.
For the chosen diagram:
- Identify its 3–4 sub-components (e.g. if the overview shows "4 Phases", drill into each phase individually)
- Generate one diagram per sub-component
- Each diagram shows the sub-component's internal workings using callout annotations — arrows pointing to specific elements with labels explaining what they are and what they do
- Keep the same character design across all Level 2 diagrams for consistency (lock the character description in every prompt)
Level 2 prompt addition:
DRILL-DOWN diagram for [COMPONENT NAME] — [TITLE]. Large central scene showing [CHARACTER] actively [DOING THE ACTION]. Multiple hand-drawn callout arrows pointing outward from different parts of the scene, each with a handwritten label describing what that element does: "[action]", "[detail]", "[outcome]". Feels like a zoomed-in explanation of one piece of a larger system.Step 1: Identify Visual Moments
For Level 1, identify 3–5 overview moments — narrative arc first:
| Type | When to use |
| Before / After | Problem → solution, chaos vs order |
| Flow / Process | Sequential steps with labeled transitions |
| Transformation | Multi-panel story showing change over time |
| Components / Anatomy | What a system is made of |
| Loop / Cycle | Repeating systems, feedback loops |
| Timeline | Phases or milestones over time |
For each moment: title, type, key elements (3–6), layout, color (red = problem, amber = process, green = solution).
Present plan then proceed immediately:
Visual moments:
1. [Title] — [type] — [elements] — [color]
2. ...
Generating now.Step 2: Write Prompts
Always include this style base:
hand-drawn educational diagram, dark charcoal background, rough sketch ink line quality, slightly imperfect hand-drawn lines, all text in handwritten chalk lettering style, warm off-white sketch lines, subtle paper texture, friendly and approachableColor palette — use purposefully:
- Warm red — problem state, chaos, the "before"
- Amber/gold — process, action, the "during"
- Green — clean output, solution, the "after"
Always add:
- Labeled arrows — label every connector with an action verb (
fires →,consolidates →,results in →) - Handwritten text — ALL labels, titles, captions look hand-lettered, slightly imperfect
- Contrasting panel moods — problem panels dense/chaotic, solution panels open/calm
- A character — include a simple sketch robot or figure for warmth; keep description identical across all diagrams in a set for consistency
Consistent robot character description (use this exact wording every time):
cute hand-drawn robot character, round head, simple antenna, small glowing chest panel, friendly expression, rough sketch ink styleFull prompt template:
hand-drawn educational diagram, dark charcoal background, rough sketch ink line quality, slightly imperfect hand-drawn lines, all text in handwritten chalk lettering style, warm off-white lines, friendly — [DIAGRAM TYPE]: [TITLE]. [PANEL/SCENE DESCRIPTIONS with color-coded borders]. [LABELED ARROWS]. [CHARACTER if relevant — use consistent description]. [BOTTOM CAPTION in handwritten font]. Spacious layout, breathing room.Example — transformation (Level 1):
hand-drawn educational diagram, dark charcoal background, rough sketch ink line quality, all text in handwritten chalk lettering — TRANSFORMATION: What the Onboarding Flow Does. Three panels connected by thick hand-drawn arrows. Panel 1 red border 'New Signup': messy overlapping to-do cards, a confused customer. Arrow labeled 'flow starts →'. Panel 2 amber border 'Onboarding Flow': cute hand-drawn robot character, round head, simple antenna, small glowing chest panel, friendly expression, rough sketch ink style — guiding the customer, three rough boxes labeled 'welcome' 'setup' 'first win'. Arrow labeled 'done →'. Panel 3 green border 'Activated Customer': neat organized dashboard, checkmark, happy customer. Timeline arrow at bottom: 'Day 1' → 'Day 3' → 'Day 7'. All handwritten text.Example — drill-down (Level 2):
hand-drawn educational diagram, dark charcoal background, rough sketch ink line quality, all text in handwritten chalk lettering — DRILL-DOWN: Step 1 — Welcome. Large central scene: cute hand-drawn robot character, round head, simple antenna, small glowing chest panel, friendly expression, rough sketch ink style — greeting a new customer and reading a large book labeled 'Account Profile'. Multiple hand-drawn callout arrows pointing outward: arrow pointing to book labeled 'reads what they signed up for', arrow pointing to robot's eyes labeled 'checks their goals', arrow pointing to open pages labeled 'personalizes the welcome', arrow at bottom labeled 'making them feel understood on day one'. Amber border. Handwritten title 'STEP 1: WELCOME' at top with a wave icon. Spacious, dark background.Step 3: Generate Images
Run all images in parallel background tasks.
python3 scripts/generate_image.py \
"<full prompt>" \
"/tmp/infographic_<slug>.png" \
"16:9" \
"2K"- Run from the skill directory so the relative
scripts/path resolves (or use the absolute path togenerate_image.py). <slug>= short snake_case title (e.g.phases_overview,drill_orient)- Default:
16:9at2K - Requires a Google Gemini API key in the environment (see Requirements above)
Step 4: Return Results
**[Topic] — [Level 1 Overview / Level 2: Drill-down on X]**
1. **[Title]**

*[One sentence: the insight this teaches]*
2. ...After Level 1: offer to drill deeper on any diagram. After Level 2: offer to drill into another component or regenerate any image.
Visual Style Reference
Use:
- Dark charcoal background
- White + warm off-white sketch lines
- Red / amber / green as story-driven accent colors
- Rough, slightly imperfect line quality
- Handwritten text for ALL labels, titles, captions
- Labeled arrows (never silent)
- Contrasting panel density (chaotic before, calm after)
- Consistent robot character across all images in a set
Avoid:
- Clean flat vector / computer-perfect lines
- Silent unlabeled arrows
- Monochromatic layouts
- Clutter — max 6 elements per panel
- Typed/print fonts anywhere
- Inconsistent character appearance across a set
Troubleshooting
- Text looks typed: Emphasize "all text handwritten chalk lettering, hand-lettered, slightly imperfect"
- Too monochromatic: Name border colors explicitly per panel
- Arrows not labeled: Add each arrow label explicitly in the prompt
- Character looks different across images: Use the exact same robot character description verbatim in every prompt
- Script errors:
pip install google-genai Pillow TypeError: 'NoneType' object is not iterable/ no image returned: Gemini returnedMALFORMED_FUNCTION_CALL. Two known causes:- "callout card" in prompt — model interprets it as a function call instruction. Always frame prompts as "hand-drawn educational diagram", never "callout card".
- Slash commands are fine — writing
/tasks,/memoryetc. literally in prompts does not cause this error. The root cause is always "callout card" framing or shell¶llelism.
- Parallel generation failures: Don't run multiple scripts via shell
&loops. Use Claude Code's background task system (run_in_background: true) to run all images in parallel reliably. - Wrong model: Only use
gemini-3.1-flash-image-preview. Do not use older or alternative model IDs.
Cue / Command Slides
For slides that show a screen command the creator should type (e.g. /memory, /tasks), use this framing:
hand-drawn educational diagram, dark charcoal background, rough sketch ink line quality, all text in handwritten chalk lettering — SCREEN CUE SLIDE: [command]. Large centered text 'forward-slash [command]' in bold handwritten lettering, warm amber glow. [Supporting label]. Dark background, minimal, high contrast.The different visual styling (large centered command, amber glow) signals the cue — no label like "SWITCH TO CLAUDE CODE" needed.
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
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.

