Summary
This validates your agent skill definitions against the agentskills.io spec and AGENTS.md rules before you commit them. Run it when you're creating or modifying skills to catch missing fields, malformed YAML, or incomplete documentation. It's purely a contract validator, not a logic tester, so it won't tell you if your skill actually works, just whether it's properly structured and machine-readable. The real value is in CI/CD pipelines where you can block non-compliant PRs automatically. It only needs read access and runs everything locally without external calls, which makes it safe to run on any repo without worrying about data leaks.
Install to Claude Code
npx -y skills add jorgealves/agent_skills --skill skill-validator --agent claude-codeInstalls into .claude/skills of the current project.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Files
SKILL.mdView on GitHub
Skill Validator
Purpose and Intent
The skill-validator is the primary quality control tool for this repository. It ensures that every AI Agent Skill is "machine-consumable" by validating it against the agentskills.io standard and the repository's AGENTS.md rules.
When to Use
- During Development: Run this skill every time you create or modify a skill.
- CI/CD Integration: Automatically block Pull Requests that contain non-compliant skill definitions.
- Repository Audits: Periodically scan the entire
agent-skills/directory to ensure long-term compliance as specifications evolve.
When NOT to Use
- Code Logic Validation: This tool does not "run" the skills or verify their internal logic; it only validates the "contract" (the YAML and Markdown).
Error Conditions and Edge Cases
- Invalid YAML: If
skill.yamlcannot be parsed, the validator will fail immediately with a syntax error. - Missing Required Fields: Any missing field defined in the
agentskills.iospecification will result inis_valid: false. - Empty Directories: Folders in
agent-skills/that do not contain askill.yamlwill be flagged as invalid skills.
Security and Data-Handling Considerations
- ReadOnly Access: The skill only requires read permissions for the target directory.
- No External Calls: Validation is performed locally using the embedded schema rules.
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
