BuildForPeople

LiveOpen source

Skill Vision

A Claude Agent Skill validator for Claude Code: spec conformance, script testing, quality and security scoring, a Mermaid skill map and per-skill token cost.


Skill Vision is a Claude Agent Skill validator: an Agent Skill for Claude Code that inspects your other skills before they ship. Ask Claude "validate my skill" and it tells you what would keep the skill from loading, uploading or triggering.

  • Spec conformance, by the letter of the Agent Skills spec, with each skill's token cost: description tokens every session, body tokens on trigger.
  • Script testing, recursive, with a timeout so a hanging script cannot hang the check.
  • Quality and security scores, with a grade from A+ to F and an ordered list of what to fix first.
  • Documentation truth: a claim auditor checks the docs against the code, including "runs offline" claims.
  • A Mermaid map of everything Claude can reach from SKILL.md.

The scripts use only the Python standard library, open no network connection and print JSON for CI. 196 adversarial tests cover them. Part of a three-skill suite with Research Hound and Agent Oracle.

Read more: Skill Vision: validate your Claude AI skills before they ship.

Questions about Skill Vision?

Ask about features, licensing, pricing or setup. Your email goes to [email protected] and the reply comes back to your inbox.

Email a question