Auditing Agent Config — Security skill for Claude Code
Audit your Claude setup for drift: redundant skills, dead plugins, bloated CLAUDE.md, session-start pile-up.
How to install Auditing Agent Config
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open keshavjoglekar/auditing-agent-config and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Auditing Agent Config does
Audit your Claude setup for drift: redundant skills, dead plugins, bloated CLAUDE.md, session-start pile-up.
Alternatives in Security
- Reflexion — Audit and optimize the .claude/ ecosystem — skills, hooks, guides, patterns, agents, and settings — for qualit 224 ★
- AWS MCP Server — by alexei-led - Features multiple Python environment setup options with detailed code style guidelines, compre 182 ★
- Sovereign Skills — 20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session man 125 ★
README
Auditing Agent Config
A portable agent skill that audits your Claude setup for drift — redundant skills, forgotten plugins, bloated CLAUDE.md files, session-start pile-up, and MCP servers nobody uses. It reports; it doesn't touch anything unless you tell it to.
It is plain Markdown, so it runs in any harness that supports skill-style instructions.
Installation
Skills CLI
npx skills add keshavjoglekar/auditing-agent-config
Install globally, across every project:
npx skills add keshavjoglekar/auditing-agent-config -g
Claude Code plugin
/plugin marketplace add keshavjoglekar/auditing-agent-config
/plugin install auditing-agent-config@auditing-agent-config
Claude desktop app / claude.ai
Download [`auditing-agent-config.zip`](https://github.com/keshavjoglekar/auditing-agent-config/releases/latest/download/auditing-agent-config.zip), then Settings → Capabilities → Skills → upload.
Manual
mkdir -p ~/.claude/skills/auditing-agent-config
curl -fsSL https://raw.githubusercontent.com/keshavjoglekar/auditing-agent-config/main/SKILL.md \
-o ~/.claude/skills/auditing-agent-config/SKILL.md
Restart your agent afterward. Skills load at session start.
Usage
Audit my Claude setup. Report only, don't change anything.
It gathers what it can reach, asks you to paste output from the interactive commands it can't run itself, and returns a report: findings ranked by impact, then suggested improvements, then what it checked and found fine.
Re-run it every few months, before installing something that might overlap what you have, or whenever a skill stops triggering when you expect it to.
Why config drift happens
Skill libraries accumulate cruft exactly like codebases do. Nobody removes the old thing when they add the thing that replaced it. Six months in, you have four tools that all claim to design UIs, a plugin firing a large instruction block at the start of every session, and no idea which of thos
Related Skills
Setup Dead Weight Audit
Audit Claude setup files for dead-weight instructions and config entries that would not change output or behav
Start Audit
Start an AUDIT pane session — adversarial review + prompt-master directives
Start Impl
Start an IMPL pane session — read context and execute AUDIT directives
SEO Audit Multilocale
Use when auditing a multi-language website (Indonesian + English, plus any other locales) for SEO correctness
Review Code Maintainability
Structural maintainability audit: god objects, hidden coupling, dead code, comment-standard compliance, and dr
Audit Session
Find your own most-violated CLAUDE.md rules by auditing recent session transcripts
Related Agents
Setup Guide
Run automatically at the start of every Lumen session to validate MCP connections and seed the context object.
Project Auditor
Read-only health audit of the VibeFrame monorepo. Use for a periodic project review — structure, CLI/doc drift
API Cost Scanner
Scans codebase for AI/API cost optimization opportunities including model usage, token estimates, caching stra