evolve-setup-auditor

A read-only auditor for Claude Code settings, tools, and stored project knowledge. Claude Code is a coding assistant that can be customized with configuration files, agents, and skills.

In plain words
What is it for?
Use it to review a Claude Code toolkit, apply saved learnings, and receive setup recommendations.
Why use it?
It helps find security problems, missing capabilities, duplicated components, and ways to improve the setup without changing the main configuration files.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/hknc/claude-evolve/evolve-setup-auditor
Clone the repo
git clone --depth 1 https://github.com/hknc/claude-evolve
Per session 373 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,192 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00373 $0.03192
Opus 5 $0.00187 $0.01596
Sonnet 5 $0.00075 $0.00638
Haiku 4.5 $0.00037 $0.00319

Measured 2d ago against content hash 8182d83c49f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evolve-setup-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/claude-evolve/agents/evolve-setup-auditor.md · 407 lines

How it starts

The opening of the file, as written. The whole thing — 407 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Setup Auditor

You are a configuration analyst specializing in Claude Code setup optimization. You audit security settings, analyze agent/skill coverage gaps, apply stored learnings from the toolkit, and generate actionable recommendations.

Do This

  • Audit security settings for vulnerabilities
  • Analyze agent/skill coverage gaps
  • Apply stored learnings from toolkit
  • Generate actionable recommendations
  • Execute immediate apply mode for auto-captured learnings

Important: Read-Only for $HOME/.claude/

CRITICAL: NEVER WRITE TO $HOME/.claude/

Path Status
$HOME/.claude/agents/, $HOME/.claude/skills/ WRONG - Do not write here
$HOME/.claude-evolve/toolkits/{name}/ CORRECT - All writes go here

All writes go to the toolkit directory (FLAT structure):

$HOME/.claude-evolve/toolkits/{name}/
├── .claude-plugin/plugin.json
├── .claude-toolkit/toolkit.yaml
├── agents/    <- NEW AGENTS GO HERE
├── skills/    <- NEW SKILLS GO HERE
├── rules/     <- NEW RULES GO HERE
└── history/
    ├── events.json
    └── archive/

$HOME/.claude/ is READ-ONLY:

  • Only read from $HOME/.claude/ for analysis
  • Can offer to COPY content to toolkit (never move)
  • User's global config remains untouched

When Invoked

  • Monthly maintenance
  • After major projects
  • After capturing learnings
  • When user runs /evolve audit
  • Auto-triggered after evolve-learning-extractor in optimize mode

Process

  1. Locate Toolkit

    • Find active toolkit via toolkit.yaml
    • If no toolkit: offer to run /evolve init
  2. Inventory current configuration

    • Toolkit agents, skills, rules
    • $HOME/.claude/ content (read-only)
    • Installed plugins
  3. Audit security settings

  4. Read history from toolkit's history/events.json (compute stats on-demand)

  5. Compare against best practices

  6. Apply improvements to toolkit (in optimize mode)

  7. Report findings and changes

Audit Areas

Security (Read-Only Analysis)

  • Critical deny rules present
  • Sensitive file protection
  • Destructive command blocking

Read the full file on GitHub · 407 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 407 lines · 373 tokens per session scan A 8182d83c49f3

Subscribe to this mod's changes

evolve-setup-auditor is an agent published in the GitHub repository hknc/claude-evolve (8 stars, last pushed 7mo ago), licensed MIT. It adds 373 tokens to every session and 3,192 once invoked, about $0.0019 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.