skills.rest
JaimeJunr
JaimeJunrCOMMUNITY
Explore all skills by this author →
GitHub

harness-audit

Audit Claude Code configurations and generate a prioritized quality scorecard.

2|Updated Apr 11, 2026
Search Tags:#security#audit#hooks#quality#best-practices#claude-code#scorecard
One-click install
npx skills add https://github.com/JaimeJunr/context-mode --skill harness-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harness-audit
Source: https://github.com/JaimeJunr/context-mode/tree/main/.claude/skills/harness-audit
Command: npx skills add https://github.com/JaimeJunr/context-mode --skill harness-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Esta Skill executa uma auditoria da configuração Claude Code no repositório atual e retorna um scorecard de qualidade organizado em 7 categorias, identificando lacunas e sugerindo melhorias prioritárias.

Core Features & Use Cases

  • Inventário de componentes (skills, commands, hooks, e MCP servers) e verificação de CLAUDE.md, hooks e gates.
  • Geração de scorecard detalhado com ações prioritárias para correções, melhorias de segurança e eficiência de contexto.
  • Recomenda recomendações acionáveis para equipes de engenharia, com foco em redução de tokens e melhoria de conformidade.

Quick Start

Run /harness-audit with an optional scope to generate a prioritized quality scorecard for the current Claude Code setup.

Frequently Asked Questions about harness-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit my Claude Code setup for quality and security gaps?▼

Auditing Claude Code configurations involves checking CLAUDE.md, hooks, quality gates, and MCP servers to generate a detailed scorecard. This scorecard identifies gaps in tool coverage, context efficiency, memory persistence, and eval coverage with actionable recommendations.

What does a Claude Code quality scorecard evaluate?▼

The scorecard evaluates configurations by inventorying components like skills, commands, hooks, and MCP servers. It checks for CLAUDE.md presence, hook implementations, and quality gates to produce actionable, priority-ranked corrections for security and token reduction.

How do I check if my Claude Code hooks and quality gates are configured correctly?▼

To verify hooks and quality gates, run an audit that inspects the current repository setup for component inventory and gate verification. The resulting scorecard provides prioritized actions to correct missing hooks and improve quality enforcement.

Can I audit Claude Code MCP servers and commands for token efficiency?▼

Auditing Claude Code configurations includes inventorying MCP servers and commands to assess context efficiency and cost efficiency. The generated scorecard provides actionable recommendations focused on token reduction and improved setup conformity for engineering teams.

What is the best way to improve Claude Code security guardrails and eval coverage?▼

Improving security guardrails and eval coverage involves auditing your Claude Code configuration to generate a prioritized scorecard. This scorecard highlights missing security checks and evaluation gaps, providing actionable recommendations for engineering teams to implement.

Do I need any dependencies to run a Claude Code configuration audit?▼

You do not need any dependencies to perform a Claude Code configuration audit. The audit tool works standalone to evaluate your repository setup, including hooks, skills, and agents, generating a comprehensive quality scorecard with prioritized actions.

Related Skills

Explore More Software Engineering Skills→
carlo-spadacarlo-spada

claude-code-features

Explain Claude Code extensibility features including skills, slash commands, hooks, MCP servers, and CLAUDE.md.

Community
Basic
MichaelvanLaarMichaelvanLaar
1

cc-optimize

Audit Claude Code configurations and generate a prioritized remediation plan.

Community
Advanced
AsiaOstrichAsiaOstrich
70

checkin

Verify build status, tests, code quality, and workflow compliance before committing.

Community
Intermediate
flonatflonat
127

code-review

Review R and Python research scripts against an 11-category scorecard.

Community
Advanced