WebDofollow
  • Submit your website
  • AI Tools
  • Rankings
  • Skills
  • Prompts
  • Blog
  • About
Back to prompts
Codex / Claude Code·Skill creation / Security·Advanced

Agent Skill Quality Review

Review a SKILL.md for scope, determinism, safety, context cost, and testability before installation.

Complete prompt

Copy, then adapt
Review the supplied Agent Skill as an installer and evaluator. Check trigger precision, scope, required context, tool permissions, destructive actions, external data handling, output contract, failure behavior, and test coverage. Separate confirmed issues from suggestions. Return blocking issues first, then a minimal patch plan.
Copy

When to use it

Run this before installing a community Skill or publishing an internal Skill.

Review standard

The reviewer must distinguish exploitable or blocking risks from preferences. Every blocking issue needs a quoted instruction and a concrete failure mode.

Expected output

  1. Decision: approve, approve with changes, or reject.
  2. Blocking findings with evidence.
  3. Context and maintenance cost.
  4. Minimal patch plan.
  5. Evaluation cases.

Common failure

A long style critique without a reproducible failure is not a useful review.

Prompt or Skill?

If this prompt needs stable triggers, fixed tools, supporting files, and evaluation cases, validate it manually and then promote it to an Agent Skill.

WebDofollow

Explore AI tools, practical Agent Skills, prompts, and field-tested workflows.

Explore
  • Submit your website
  • AI Tool Directory
  • AI Rankings
  • Agent Skills
  • Prompt Library
Resources
  • Blog
  • Browse Categories
Company
  • About
  • Contact
Legal
  • Cookie Policy
  • Privacy Policy
  • Terms of Service
© 2026 WebDofollow. All Rights Reserved.