monoskill
CLAUDE.md0 starsChanged 4 months agoMITView on GitHub
Monoskill is a Node.js CLI that compiles a repository containing many agent skills into one provenance-aware router skill. The project is graveyard-managed: GitHub issues are the work queue, pull requests are the delivery surface, and the strategic project state is linked from docs/OVERVIEW.md. repo-type: code project-types: none
In this repository6 other files it gives its agents.
- data-visualizationSkill
- explainer-visualsSkill
- gsap-coreSkill
- gsap-performanceSkill
- gsap-timelineSkill
- motion-art-directionSkill
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
- Very specific: line 10
Read docs/ARCHITECTURE.md before changing compiler boundaries or output layout.
- Very specific: line 11
Read docs/DECISIONS.md before revisiting settled product or implementation choices.
- Very specific: line 12
Use docs/OVERVIEW.md to find the human-facing strategic state in Notion.
- Distinctive: line 13
Treat open GitHub issues as future work.
- Very specific: line 19
Run npm run check for JavaScript syntax validation.
- Distinctive: line 23
Keep source resolution compatible with GitHub shorthand, Git URLs, and local directories.
- Distinctive: line 24
Every merge must close or explicitly advance its GitHub issue and include tests proportional to the behavior changed.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 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
Monoskill agent guide
Monoskill is a Node.js CLI that compiles a repository containing many agent skills into one provenance-aware router skill. The project is graveyard-managed: GitHub issues are the work queue, pull requests are the delivery surface, and the strategic project state is linked from docs/OVERVIEW.md.
repo-type: code project-types: none
Canon
- Read
docs/ARCHITECTURE.mdbefore changing compiler boundaries or output layout. - Read
docs/DECISIONS.mdbefore revisiting settled product or implementation choices. - Use
docs/OVERVIEW.mdto find the human-facing strategic state in Notion. - Treat open GitHub issues as future work. Do not write unshipped aspirations into architecture canon.
Working rules
- Requires Node.js 20 or newer.
- Install dependencies with
npm installand run the complete test suite withnpm test. - Run
npm run checkfor JavaScript syntax validation. - Exercise CLI changes through the published-style entry point:
node bin/monoskill.js <command>. - Preserve the generated skill contract: a compact root
SKILL.md,agents/openai.yaml, full upstream trees underreferences/, andprovenance.json. - Preserve user artifacts. Never overwrite an output directory or packaged artifact silently.
- Keep source resolution compatible with GitHub shorthand, Git URLs, and local directories.
- Every merge must close or explicitly advance its GitHub issue and include tests proportional to the behavior changed.
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.

