agentleFS
Sign inStart free
kimasplundkimasplund/thinkbetter-skills/.claude/CLAUDE.md

thinkbetter-skills

CLAUDE.md0 starsChanged 32 days agoMITView on GitHub

The file

We found no rules to list in this file: it reads as prose or reference rather than instructions.

Create a free account to mount
Files
CLAUDE.md

thinkbetter-skills: repo rules

  • Skills ship from skills/<name>/, and nothing else in the repo is installed. Validation records go in evolution/<skill>/ and example outputs in examples/<skill>/. Never put either inside a skill directory.
  • A skill's text changes only through a pre-registered skill-gauntlet round, or through an owner decision disclosed in the README and CHANGELOG. Mechanical fixes such as YAML quoting, broken links or formatting are fine without a round, but prove the content is unchanged.
  • Every content change bumps that skill's metadata.version, the plugin version in .claude-plugin/plugin.json, and adds a CHANGELOG.md entry. Never set version in marketplace.json.
  • Before committing: enable the pre-push gate once per clone with git config core.hooksPath .githooks, then run python3 -m unittest discover -s tests, python3 scripts/validate_skills.py, and claude plugin validate . --strict (plus .claude-plugin/plugin.json and ./skills).
  • Keep frontmatter to the six Agent Skills spec keys. Claude Code-only keys (model, context, agent, user-invocable) are allowed, but build_zips.py strips them for claude.ai.
  • Never commit private material. Project-specific overlays (for example a domain pack under references/) go in .git/info/exclude, not .gitignore, so even their filenames stay private. Put private terms in the local, git-ignored .privacy-denylist so the validator catches leaks.
  • This machine runs the skills through symlinks made by scripts/link_local.sh, so edits here are live immediately. Don't also install the plugin locally, or every skill appears twice.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free