Automated summary
Audits and polishes Codex and Claude skills to ensure structural integrity and standardization using dedicated tooling.
Strengths
- Automates validation of skill structure
- Supports cross-flavor audits (Codex and Claude)
- Provides deterministic polishing scripts
- Generates reports with scores
Limitations
- Requires existing Codex polisher setup
- Only works with specific directory structures
- Does not modify skills unless explicitly requested
When you need to ensure all Codex/Claude skills follow structural standards before deployment or sharing.
When you want to create or modify skills, as this is an auditing tool.
Security analysis
SafeThe skill executes only local, pre-existing scripts (uv run, PowerShell) for auditing purposes. No network downloads, arbitrary code execution, or destructive commands are instructed. Bash is allowed-tool but commands are deterministic and safe.
No concerns found
How this analysis is madeExamples
Run the Codex skill-polisher on all skills in .codex/skills and report any issues.Check all Claude skills in .claude/skills for valid SKILL.md frontmatter (name and description).Perform a cross-flavor audit on both .codex/skills and .claude/skills using the shared auditor.name: skill-polisher
description: Meta-skill for Claude Code to validate and polish Codex skills under .codex/skills/ by running the skill-polisher and reviewing structural integrity. Use when auditing Codex skills, enforcing standardization, or running integrity sweeps across the Codex skill set.
allowed-tools: "Read, Write, Glob, Grep, Bash"
user-invocable: true
Skill Polisher (Claude Code)
Audit Codex skills for structural integrity and standardization by using the existing Codex skill-polisher tooling.
Flavor Contract
- Claude-flavored by default: validates Claude frontmatter/tooling rules.
- Cross-flavored compatible via shared auditor (see Cross-Flavor Audit).
Target scope
- Codex skills:
- Single skill:
.codex/skills/<skill-name> - Whole set:
.codex/skills
- Single skill:
- Claude skills:
- Single skill:
.claude/skills/<skill-name> - Whole set:
.claude/skills
- Single skill:
Execution
Codex skills (preferred)
Use the Codex polisher for deterministic audits:
uv run .codex/skills/skill-polisher/scripts/polish_skill.py .codex/skills --all
For a single skill:
uv run .codex/skills/skill-polisher/scripts/polish_skill.py .codex/skills/<skill-name>
Claude skills (manual audit)
Claude skills do not share the Codex polisher; use a lightweight audit:
Get-ChildItem .claude/skills -Directory
Get-Content .claude/skills/<skill-name>/SKILL.md
Checks to confirm:
SKILL.mdpresent- Frontmatter includes
nameanddescription - Any
allowed-toolsentries match Claude tool names (Read,Write,Glob,Grep,Bashas needed)
Cross-Flavor Audit (shared)
Use the shared auditor when you want consistent reporting:
uv run scripts/skill_audit.py --flavor codex --root .codex/skills
uv run scripts/skill_audit.py --flavor claude --root .claude/skills
Hook (Claude-side)
Run the Claude cross-compatible audit via the helper script:
.\scripts\run_claude_skill_polisher.ps1 -Root .claude/skills
Claude-Local Audit (Claude-centric outputs)
Write JSON outputs into claude/mailbox:
.\scripts\run_claude_local_audit.ps1 -Root .claude/skills
Hook (Claude-side cross-polish)
Run both Codex and Claude audits in one pass:
.\scripts\run_claude_cross_polish.ps1 -CodexRoot .codex/skills -ClaudeRoot .claude/skills
Bridge (Codex tooling)
Invoke Codex polisher from Claude:
.\scripts\run_codex_polisher.ps1 -Root .codex/skills
Reporting
Summarize:
- TASTE + score
- Issues (if any)
- Files changed (if any)
Notes
- This skill audits Codex and Claude skills; do not modify either unless explicitly requested.
- If you need to add
@POLISHEDseals, do so after a clean run and record the timestamp.
Cross-Flavor Compatibility
- Codex flavor: requires
agents/openai.yamlandassets/with SVG icons. - Claude flavor: requires
SKILL.mdwith valid frontmatter (name,description), optionalallowed-tools. - For shared audits use:
python scripts/skill_audit.py --flavor codex --root .codex/skillsandpython scripts/skill_audit.py --flavor claude --root .claude/skills.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/skill-polisher-mma3jr3j/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/skill-polisher-mma3jr3j' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/skill-polisher-mma3jr3j/download' --output '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md.part' && mv '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md.part' '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Tools declared by the author (check availability in your agent): Read, Write, Glob, Grep, Bash
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md' && test -s '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/skill-polisher' && cp -R '.skill-downloads/skill-polisher-mma3jr3j' '.claude/skills/skill-polisher'Codex
test -f '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md' && test -s '.skill-downloads/skill-polisher-mma3jr3j/SKILL.md' && mkdir -p '.agents/skills' && test ! -e '.agents/skills/skill-polisher' && cp -R '.skill-downloads/skill-polisher-mma3jr3j' '.agents/skills/skill-polisher'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/skill-polisher-mma3jr3j'openskills read 'skill-polisher-mma3jr3j'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/skill-polisher-mma3jr3j/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.