kanban-board-mcp / hand-drawn-diagrams
CLAUDE.md1 starsChanged 4 days agoMITView on GitHub
hand-drawn-diagrams is an AI skill that converts natural language prompts into hand-drawn Excalidraw diagrams. It produces a hosted edit URL, animated SVG, and PNG — from a single prompt, no app required. This repository contains the skill definition, rendering scripts, and installer for Claude Code and compatible agents. The skill is activated via SKILL.md. When activated, it follows workflow.md:
1 other project uses a version of this.
Adopted by
Most recently taken up by
- muthuishere/hand-drawn-diagramsword for word, first seen Apr 17, 2026
Other versions1 project uses a version of this.
Word for word · 1
- muthuishere/hand-drawn-diagrams★ 78100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
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 repository24 other files it gives its agents.
- .github/skills/hand-drawn-diagrams/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- canvas-designSkill
- baSkill
- consultantSkill
- critical-thinkingSkill
- cybersecuritySkill
- debuggerSkill
- designerSkill
- doc-writerSkill
- frontend-designSkill
- git-workflowSkill
- hand-drawn-diagramsSkill
- marketingSkill
- philosopherSkill
- pmSkill
- psychologistSkill
- qcSkill
- remotion-best-practicesSkill
- setup-copilotSkill
- technical-leadSkill
- test-data-generatorSkill
- unit-testingSkill
The file
We split this CLAUDE.md into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
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 68
Write to workspace only if user specifies a path
- Very specific: line 114
Do not write .excalidraw files to the user's workspace without being asked
- Very specific: line 115
Do not run get_excalidraw_urls.py before validate_excalidraw.py exits 0
- Very specific: line 116
Do not render before the .excalidraw file is written and validated
- Very specific: line 117
Do not use excalidraw.com directly — use edit_excalidraw.py to get hosted URLs
- Distinctive: line 118
Do not add multiple colors/fills for decoration — keep diagrams monochrome by default
- Distinctive: line 119
Do not write paragraphs inside shapes — diagrams use 1–5 word labels only
≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CLAUDE.md
Project Overview
hand-drawn-diagrams is an AI skill that converts natural language prompts into hand-drawn Excalidraw diagrams. It produces a hosted edit URL, animated SVG, and PNG — from a single prompt, no app required.
This repository contains the skill definition, rendering scripts, and installer for Claude Code and compatible agents.
Repository Structure
hand-drawn-diagrams/
├── SKILL.md # Skill entry point (name + description)
├── workflow.md # Master workflow — read this first
├── steps/
│ ├── step-01-route.md # Pick diagram type from routing table
│ ├── step-02-draw.md # Generate .excalidraw + .animationinfo.json
│ └── step-03-validate.md # Validate, get hosted URL, offer animation
├── references/
│ ├── activation-routing.xml # Route selection rules
│ ├── fundamental-shapes.md # Core shape language
│ ├── json-schema.md # .excalidraw JSON schema
│ ├── element-templates.md # Grid layout + element templates
│ ├── animation-spec.md # Animation story patterns
│ ├── quality-checklist.md # Diagram quality rules
│ ├── patterns.md # Reusable layout patterns
│ └── *-diagrams.md # Per-route guides (teachers, ux, funnel…)
├── scripts/ # Python rendering tools (managed with uv)
│ ├── validate_excalidraw.py # Pre-flight validation — run before URL gen
│ ├── get_excalidraw_urls.py # Generates hosted Edit + Animate URLs
│ ├── edit_excalidraw.py # Opens diagram in hosted editor
│ ├── render_excalidraw.py # PNG export (Playwright fallback)
│ ├── render_animated_svg.py # Animated SVG export (Playwright fallback)
│ ├── animate_excalidraw.py # Animation helper
│ ├── hosted_scene_urls.py # URL encoding utilities
│ ├── local_excalidraw_server.py
│ ├── pyproject.toml # Python deps: playwright>=1.40, pytest
│ └── tests/ # pytest test suite
├── installscripts/
│ ├── install.py # Detects Claude Code / Agent CLI, installs skill
│ └── uninstall.py
├── install.sh / install.cmd # Shell wrappers → installscripts/install.py
├── uninstall.sh / uninstall.cmd
└── docs/ # Additional documentationTech Stack
- Language: Python 3.11+
- Package manager:
uv(always useuv run python ...for scripts) - Rendering: Chrome DevTools MCP (preferred) or Playwright (fallback)
- Output format:
.excalidrawJSON,.animationinfo.json,.animated.svg,.png - Diagram host: Excalidraw hosted editor (gzip/base64 scene in URL hash)
Skill Activation
The skill is activated via SKILL.md. When activated, it follows workflow.md:
- Route (
step-01-route.md) — pick one diagram type - Draw (
step-02-draw.md) — write.excalidraw+.animationinfo.json - Validate (
step-03-validate.md) — validate, generate hosted URL, deliver
Key Conventions
File locations
.excalidraw+.animationinfo.json→/tmp/hand-drawn-diagrams/<slug>/(never litter workspace).animated.svg→ always the user's project/workspace directory- Write to workspace only if user specifies a path
Rendering commands (always run from scripts/)
cd scripts
uv run python validate_excalidraw.py "/absolute/path/to/file.excalidraw"
uv run python open_diagram.py "/absolute/path/to/file.excalidraw" # preferred: writes open.html + opens browser
uv run python get_excalidraw_urls.py "/absolute/path/to/file.excalidraw" # fallback: prints raw URLs
uv run python render_excalidraw.py "/absolute/path/to/file.excalidraw"
uv run python render_animated_svg.py "/tmp/.../diagram.excalidraw" --output "/project/name.animated.svg"Rendering priority
- Chrome DevTools MCP — fast, uses real browser, no install required
- Playwright scripts — fallback only
Output delivery order
- Hosted edit URL (always first, after validation passes)
- Offer animation ("Want a video version?")
- PNG only if explicitly requested
Diagram style defaults
- Hand-drawn / sketch font
- Monochrome (no fills unless essential)
- Files go to
/tmp/— workspace stays clean
Running Tests
cd scripts
uv run pytest # all tests
uv run pytest -m "not slow" # skip Playwright/Chromium tests
uv run pytest tests/test_validate_excalidraw.pyInstallation / Uninstall
bash install.sh # macOS/Linux — detects Claude Code and Agent CLI
install.cmd # Windows
bash uninstall.sh # macOS/Linux
uninstall.cmd # WindowsWhat NOT to do
- Do not write
.excalidrawfiles to the user's workspace without being asked - Do not run
get_excalidraw_urls.pybeforevalidate_excalidraw.pyexits 0 - Do not render before the
.excalidrawfile is written and validated - Do not use
excalidraw.comdirectly — useedit_excalidraw.pyto get hosted URLs - Do not add multiple colors/fills for decoration — keep diagrams monochrome by default
- Do not write paragraphs inside shapes — diagrams use 1–5 word labels only
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.

