Borrowing it
Nothing to install: this file belongs to dzhng/duet-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dzhng/duet-agent/main/.agents/skills/audit-agents/SKILL.mdgit clone --depth 1 https://github.com/dzhng/duet-agentWrote 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/dzhng/duet-agent/audit-agents)<a href="https://agentmods.dev/skills/dzhng/duet-agent/audit-agents"><img src="https://agentmods.dev/badge/skills/dzhng/duet-agent/audit-agents/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/dzhng/duet-agent/audit-agents"><img src="https://agentmods.dev/badge/skills/dzhng/duet-agent/audit-agents.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 | $0.00036 | $0.00760 |
| Opus 5.5 | $0.00014 | $0.00304 |
| Sonnet 5.5 | $0.00007 | $0.00152 |
| Haiku 4.5 | $0.00004 | $0.00076 |
Grade A, and why
audit-agents 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 25d 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit Agents
AGENTS.md is read by everyone who will ever work on the project. Picture the next thousand contributors over the next hundred years: a line belongs only if it is still true, and still changes what they do, after every file, tool and dependency has been replaced.
Workflow
- Read the AGENTS.md and the root readme. This is a document audit: run no builds and no tests.
- Give every line one verdict, using the tests below.
- Keep: a principle that passes every test.
- Rephrase: a lasting lesson worded in today's mechanics, or too vague to act on.
- Move: a detail someone needs. Name the owner it belongs to: a readme, a manifest, a skill or a spec.
- Delete: repetition, history, or a line that changes nothing.
- Check the shape against the example below.
- Report the findings, highest impact first. Quote each line, give its verdict and the replacement text, and group repeats. Report no finding you can't quote.
- Edit only when the user asks for a rewrite. Then start from the example, write the file, and reread it as a newcomer with no history of this session.
Tests for a line
- A hundred years. No commands, flags, paths, file or function names, dependency choices, tuned values, plans, status or bug stories. Keep the lesson and move the mechanics to their owner. The one exception is a pointer to an owner (a readme or a skill), which exists so the detail can live there.
- A thousand people. A newcomer can act on it without knowing this session, this author or this month's work.
- Changes what they do. Delete a line a capable contributor already follows.
- Plain and concrete. A principle is not an abstraction. "Look at the picture" works; "match verification to the claim" doesn't. Prefer a short sentence with its consequence, and one example from the product over a general noun.
- Iteration speed. This is the core principle and must be stated outright: optimize the time to feedback you can trust, run the narrowest check that answers the question, and run everything once, when the spec's implementation is finished. Flag any rule that adds process to every loop, puts the full run before a commit, merge, push or finished feature, or names checkpoints for it along the way. The full run comes earlier only when the next piece of work can't be trusted without it. Never trade away a real acceptance requirement for speed.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 25d ago Changed · +1 lines · -13 tokens per session 78054e2a2ada
- 26d ago First seen · 41 lines · 49 tokens per session scan A 0e4288a0a6ac
audit-agents is a skill published in the GitHub repository dzhng/duet-agent (52 stars, last pushed 7d ago), licensed Apache-2.0. It adds 36 tokens to every session and 760 once invoked, about $0.0001 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 2026-10-02.
Other skills, from other repositories
harness-creator
Build, audit, and improve harnesses that make AI coding agents reliable: AGENTS.md/CLAUDE.md instruction files, feature/state tracking, verification gates, scope boundaries, session handoff, memory persistence, context budgets, tool-permission safety, and multi-agent coordination. Use this whenever a coding agent is…
evaluator-write-qa-parallel
Internal Auto-Harness evaluator skill for parallel sprint QA and QA report writing. Use only inside the Evaluator subagent during evaluatorqaparallel.
evaluator-write-qa
Internal Auto-Harness evaluator skill for sprint QA and QA report writing. Use only inside the Evaluator subagent during qa mode.
evaluator-write-retest-parallel
Internal Auto-Harness evaluator skill for parallel sprint retest and retest report writing. Use only inside the Evaluator subagent during evaluatorretestparallel.
evaluator-write-retest
Internal Auto-Harness evaluator skill for sprint retest and retest report writing. Use only inside the Evaluator subagent during retest mode.
generator-fix-integrator
Internal Auto-Harness generator skill for worktree-merge fix integration. Use only inside the parallel Generator integrator when it is addressing named defects from QA or retest.