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.
npx skills add ngreenwall/claude-skills --skill qa-auditgit clone --depth 1 https://github.com/ngreenwall/claude-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ngreenwall/claude-skills/qa-audit)<a href="https://agentmods.dev/skills/ngreenwall/claude-skills/qa-audit"><img src="https://agentmods.dev/badge/skills/ngreenwall/claude-skills/qa-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ngreenwall/claude-skills/qa-audit"><img src="https://agentmods.dev/badge/skills/ngreenwall/claude-skills/qa-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00156 | $0.01591 |
| Opus 5.5 | $0.00062 | $0.00636 |
| Sonnet 5.5 | $0.00031 | $0.00318 |
| Haiku 4.5 | $0.00016 | $0.00159 |
Grade A, and why
qa-audit 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: QA Audit
A static-analysis QA pass over instruction and config prose. Recognized targets: SKILL.md files, CLAUDE.md, your global context file (the standing instructions your AI tool loads every session, e.g. ~/.claude/CLAUDE.md), README.md, slash command files (.claude/commands/*.md), and other instruction/config prose files. It never executes or evaluates the target against real prompts, it reads it and reasons about it.
See the frontmatter above for what this is NOT (drift-check, skill-creator's eval feature, /code-review). Closest neighbor: token-audit reduces load cost (rewrite/split/merge for token bloat); this skill checks correctness (contradictions, dead logic, unstated dependencies).
PROTOCOL
Step 1: Resolve target(s) and guard against code
Identify the file(s) to QA.
Before doing anything else, check whether the target looks like code: it's a code file if the extension is a recognized code/data extension (e.g. .ts, .tsx, .js, .jsx, .py, .rb, .go, .rs, .java, .c, .cpp, .sh, .sql). If so, stop immediately, don't spawn any agent, and tell the user in one line: "That's a code file, not instruction/config prose, use /code-review for that."
If the extension is .md/.mdc and the path is one of the recognized instruction/config locations (a SKILL.md inside a skills folder, CLAUDE.md, your global context file, README.md, .claude/commands/*.md), proceed directly.
For anything else, an unrecognized .md location, .json, .yaml, .toml, .env, or similar config file, don't auto-reject it as code, ask the user to confirm it's instruction/config prose before proceeding.
If the file is managed by a sync tool that generates copies elsewhere (e.g. a tool that regenerates .claude/ or .cursor/ files from a single source of truth), always resolve to that source file, never a generated copy, editing a generated copy gets silently overwritten on the next sync run. Skip this check if you don't use such a tool, most projects edit SKILL.md/CLAUDE.md directly.
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.
- 2mo ago First seen · 94 lines · 156 tokens per session scan A e3ad8aec166c
qa-audit is a skill published in the GitHub repository ngreenwall/claude-skills (2 stars, last pushed 2mo ago), licensed MIT. It adds 156 tokens to every session and 1,591 once invoked, about $0.0006 per session on Opus 5.5. 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.
Other skills, from other repositories
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
design-review
Five-dimension design audit (frontend UI/UX · server data-model & flow · endpoint spec · daemon-runtime/SDK & local-cache-first · built-in skill/agent-role), executed by a NON-implementing agent. Every dimension gets a verdict; N/A only with a stated reason. A dimension the design silently skipped must be surfaced as…
codebase-design
Design maintainable module interfaces, review scoped code changes, and simplify implementations with behavioral verification. Use for code review, cleanup, testability, and module design work.
review-work
Post-implementation gate review: run manual QA on the real surface yourself, then launch ONE gate reviewer (never a panel) to audit goal, constraints, code quality, security, missed context, and QA evidence. Use before a PR handoff or when the user explicitly asks to review completed work.
tech-lead/code-review
A checklist-based method for examining code for correctness, security, performance, clarity, and ease of maintenance.
pr-blocker-summarizer
Summarizes open pull requests into a blockers-first standup digest. Activates when the user asks to summarize open PRs, find blocked pull requests, generate a PR standup, or triage review backlog from a PR export.