tzurot-skills-guide
Writing Claude Code Skills - Best Practices
Use this skill when: Creating a new skill, updating an existing skill, or reviewing skill quality.
Core Principle: Progressive Disclosure
"Don't tell Claude all the information you could possibly want it to know. Rather, tell it how to find important information so that it can use it only when needed." — Anthropic Engineering
Skills should be concise entry points, not exhaustive documentation.
Size Limits
| Metric | Target | Maximum | Action if Exceeded |
|---|---|---|---|
| Skill lines | <300 | 400 | Split or reference docs |
| CLAUDE.md | <400 | 500 | Move content to skills |
| Description char | N/A | 15,000 | Trim descriptions |
More from lbds137/tzurot
tzurot-council-mcp
Multi-perspective AI consultation. Invoke with /tzurot-council-mcp for major refactors (>500 lines), structured debugging after failed attempts, or when a technical decision has multiple viable approaches.
17tzurot-testing
Testing procedures. Invoke with /tzurot-testing for test execution, coverage audits, and debugging test failures.
16tzurot-deployment
Railway deployment procedures. Invoke with /tzurot-deployment for deploying, checking logs, and troubleshooting.
16tzurot-docs
Session workflow procedures. Invoke with /tzurot-docs for session start/end, CURRENT.md/BACKLOG.md management.
14tzurot-git-workflow
Git workflow procedures. Invoke with /tzurot-git-workflow for commit, PR, and release procedures.
13tzurot-tooling
MUST invoke when: adding CLI commands, creating dev scripts, or using pnpm ops commands. Contains ops CLI reference and script organization.
1