SkillGild
Sign inCreate account

Free tool

SKILL.md checker

Paste a skill to validate it against the Agent Skills specification and Anthropic's authoring best practices. Then run an AI review that scores whether Claude Code, Codex or Cursor will actually load it.

The rule checks run in your browser; nothing you paste is stored.

0 errors, 2 warnings

  • Warning description does not say when to use the skill

    Agents load a skill by matching the request against its description. Add a trigger such as "Use when the user asks to…" with the words people actually type. Best practices

  • Warning description is very short

    Short descriptions such as "Helps with PDFs" rarely trigger reliably. Name the tasks, file types and keywords the skill covers. Spec

  • Note Folder name must match

    Save this file as pdf-processing/SKILL.md: the spec requires name to match the folder that holds it. Spec

AI review

Scores how reliably the description will make an agent load the skill, and how usable the instructions are, with fixes for anything weak.

What it checks

Spec rules (errors)

  • YAML frontmatter between --- lines
  • name: 1 to 64 characters, lowercase letters, numbers and single hyphens, no "anthropic" or "claude"
  • description: 1 to 1,024 characters, no XML tags
  • compatibility up to 500 characters; metadata maps keys to text

Best practices (warnings)

  • The description says when to use the skill
  • Third-person description, specific rather than vague
  • Body under 500 lines and about 5,000 tokens
  • Relative paths with forward slashes

The AI review

The review uses TypeSafe's Jev model to score six things: whether the description tells an agent when to load the skill, whether it covers both what and when, how specific it is, and whether the instructions are actionable, concise and consistent with the description. Add a test request to see if an agent would load the skill for it.

1 review is free without an account, a free account includes 3 a month, and paid and Pro plans include unlimited reviews.

Questions

What does the SKILL.md checker check?

It validates the YAML frontmatter against the Agent Skills specification: name is 1 to 64 lowercase letters, numbers and single hyphens without reserved words, description is 1 to 1,024 characters, and optional fields such as compatibility and metadata have the right shape. It also flags best-practice problems from Anthropic's authoring guide, such as a description that never says when to use the skill, first-person wording, or a body over 500 lines.

Why is my Claude skill not triggering?

An agent decides whether to load a skill from its name and description alone. A description that names a broad topic, such as "Helps with PDFs", without saying when to use it rarely matches real requests. Describe what the skill does and when to use it, in the words users actually type, and keep it in the third person. The AI review scores exactly this and can test a sample request.

Is the checker free?

The rule checks are free and unlimited, and run in your browser without storing anything. The AI review gives you 1 free run without an account, 3 a month with a free account, and unlimited runs on a paid or Pro plan.

Does it work for Codex, Cursor and Gemini CLI skills?

Yes. Claude Code, Codex, Cursor and Gemini CLI all read the same SKILL.md format defined by the Agent Skills specification, so the same rules apply. Some clients read extra frontmatter fields; the checker notes those without failing the file.

What happens to the SKILL.md I paste?

The rule checks never leave your browser. When you run the AI review, the name, description and up to 8,000 characters of instructions are sent to TypeSafe's Jev model for scoring and are not stored by SkillGild; only a count of reviews is kept to apply the limits.

Next steps