audit

A read-only security audit for repositories containing AI-agent add-ons and ordinary code. It looks for backdoors, hidden instructions, persistence mechanisms, stolen credentials, data leaks, supply-chain risks, and dangerous settings.

In plain words
What is it for?
Use it to inspect skills, agents, hooks, tool servers, slash commands, plugins, and code without running the repository or changing it. Findings must include the file location and supporting text.
Why use it?
It helps you judge whether a third-party repository is safe before trusting or installing it.

Skill for Claude CodeCodex

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 skills/secawa-com/plugin-auditor/audit
Any agent
npx skills add secawa-com/plugin-auditor --skill audit
Clone the repo
git clone --depth 1 https://github.com/secawa-com/plugin-auditor

Made for: Claude Code, Codex.

Per session 76 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,790 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00076 $0.04790
Opus 5 $0.00038 $0.02395
Sonnet 5 $0.00015 $0.00958
Haiku 4.5 $0.00008 $0.00479

Measured yesterday against content hash 584a1b6dd75c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

audit scanned grade B with 1 finding 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 yesterday.

The scan reads SKILL.md. This mod also ships 9 executable files (scripts/cleanup_clone.sh, scripts/clone_repo.sh, scripts/compute_delta.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Recursive force deletemediumDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

The script validates that the path is inside `${PWD}/.plugin-auditor-tmp/` and refuses anything else — do not bypass it with `rm -rf`.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

skills/audit/SKILL.md · 277 lines

How it starts

The opening of the file, as written. The whole thing — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plugin Auditor

You are the orchestrator of a static, read-only security audit of a third-party repository. The user is about to install or has just cloned a project that contains Claude Code artifacts (skills, agents, hooks, plugins, MCP servers, slash commands) and/or general code, and they want an evidence-backed verdict before they trust it.

Hard rules

  1. Never execute audited code. No npm install, no pip install, no bash setup.sh, no make, no Docker build, no running of any binary or script that originates from the audited repository. Static analysis only.
  2. Never modify the audited repository. Read, grep, and reason — that is all.
  3. Never write outside ~/.claude/plugin-auditor-reports/ and ${PWD}/.plugin-auditor-tmp/. No system-wide changes, no dotfile edits.
  4. Every finding must cite evidence. A flag without a path:line and a quoted excerpt is not a finding — it is speculation. Drop it.
  5. When in doubt, classify as CAUTION, never OK. Silence is not safety.

Step 0 — Parse the argument

The skill receives $ARGUMENTS. Resolve it as follows:

Argument shape Action
Empty Use the current working directory as REPO_PATH.
Existing local path Use it directly as REPO_PATH.
https://github.com/... or https://gitlab.com/... (optionally with @ref) Run bash "${CLAUDE_PLUGIN_ROOT}/skills/audit/scripts/clone_repo.sh" <url> to shallow-clone into ${PWD}/.plugin-auditor-tmp/{repo}-{timestamp}/. Use the printed path as REPO_PATH. The clone lands in CWD (not /tmp/) so sub-agents can Read/Grep/Glob it without extending permissions.additionalDirectories.
Anything else Stop and report: "input could not be parsed as path or supported URL".

If the argument string contains the literal flag --delta anywhere, set DELTA=true and strip it from the path/URL before parsing.

Step 1 — Setup

Run once at the start of every audit:

bash "${CLAUDE_PLUGIN_ROOT}/skills/audit/scripts/ensure_reports_dir.sh"

Read the full file on GitHub · 277 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. yesterday First seen · 277 lines · 76 tokens per session scan B 584a1b6dd75c

Subscribe to this mod's changes

audit is a skill published in the GitHub repository secawa-com/plugin-auditor (2 stars, last pushed 1mo ago), licensed MIT. It adds 76 tokens to every session and 4,790 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens