Skill Claude CodeCodex
Audit and clean up Claude Code configuration. Checks skills, hooks, plugins, configs, commands, and stale data for issues and fixes them.
2 16d ago A 33 tokens
original MIT
2Stars on the repository
20Mods indexed here, across every type
16d agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
five-steps
02Skill Claude CodeCodex
Apply the 5-step optimization framework (question, delete, simplify, accelerate, automate—in that order) as a thinking tool for shaping work. Use it when creating new plans and when reviewing existing systems, processes, tools, and products you own. Triggered by "run this through the five steps", "5 steps this"…
2 16d ago A 91 tokens
original MIT
Skill Claude CodeCodex
Replace vague “flaky” or “flakiness” language with the narrowest evidence-supported classification of a test, build, deployment, performance, security, or CI failure. Use when triaging an intermittent, recurring, or apparently nondeterministic failure; responding to a test alert; reviewing a failure report; or someone…
2 16d ago A 77 tokens
original MIT
Skill Claude CodeCodex
Audit any JS-heavy codebase for performance issues. Checks bundle weight, code splitting, selector memoization, rendering architecture, and performance infrastructure. Produces a findings report with severity ratings and recommendations.
2 16d ago A 45 tokens
original MIT
kb-to-spec
05Skill Claude CodeCodex
Promote a private Markdown plan/note into a shareable spec committed to a GitHub repo. De-privatizes the content, writes a standalone spec with an Authors header, links that spec from the top of the private note, and keeps the two independent (no duplicated content). Use when publishing a plan as a shared design doc…
2 16d ago A 110 tokens
original MIT
Skill Claude CodeCodex
Verify a login-gated UI loads and behaves correctly, to validate product changes fast, by logging in once and attaching a headless browser over CDP for every check after. Covers the auth-once session model, a reference Playwright runner, 2x screenshots, and the collision and auth-capture gotchas that waste the most…
2 16d ago C 94 tokens
original MIT
unit-tests
07Skill Claude CodeCodex
Test quality auditor that reviews existing test suites—audits, deletes, rewrites, and fills genuine gaps. Use when asked to review tests, improve test quality, or audit a test suite. Does NOT blindly add tests.
2 16d ago A 48 tokens
original MIT
Skill Claude CodeCodex
Capture before/after screenshots for PRs with visual UI changes using Playwright.
2 16d ago A 21 tokens
original MIT