skill-validator_skill

Overview

This skill validates AI agent skill definitions against agentskills.io standards and the project’s AGENTS.md rules. It ensures each skill is machine-readable and documentation-complete before merging or release. The validator flags syntax errors, missing fields, and empty skill folders to enforce consistent, consumable skill contracts.

How this skill works

The validator parses each skill's YAML and Markdown, checks the YAML structure against the embedded agentskills.io schema, and verifies required documentation fields from AGENTS.md. Validation runs locally with read-only access and no external network calls. Results include structured pass/fail output and clear error messages for syntax, schema, or missing-file issues.

When to use it

• During development when creating or updating a skill to catch issues early.
• In CI/CD pipelines to block pull requests that contain non-compliant skills.
• As part of periodic audits to ensure long-term compliance across all skills.
• Before publishing or distributing an agent bundle to ensure machine consumption.

Best practices

• Run the validator locally before committing changes to reduce CI churn.
• Keep skill YAML minimal and conformant to the schema; avoid undocumented fields.
• Include required documentation fields in Markdown so validators can verify completeness.
• Treat validator warnings as actionable items: fix schema and documentation gaps immediately.
• Integrate the tool into pre-commit hooks or pipeline stages for consistent enforcement.

Example use cases

• A developer adds a new skill folder and runs the validator to catch YAML syntax errors.
• A CI workflow runs the validator to automatically reject PRs with missing required fields.
• A maintainer scans all skills after a specification update to find outdated definitions.
• A release process verifies every skill is machine-readable before packaging an agent distribution.