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 Kevin-Liu-01/Agent-Machines --skill exemplar-auditgit clone --depth 1 https://github.com/Kevin-Liu-01/Agent-MachinesWrote 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/kevin-liu-01/agent-machines/exemplar-audit)<a href="https://agentmods.dev/skills/kevin-liu-01/agent-machines/exemplar-audit"><img src="https://agentmods.dev/badge/skills/kevin-liu-01/agent-machines/exemplar-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/kevin-liu-01/agent-machines/exemplar-audit"><img src="https://agentmods.dev/badge/skills/kevin-liu-01/agent-machines/exemplar-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 103 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00070 | $0.01092 |
| Opus 5 | $0.00035 | $0.00546 |
| Sonnet 5 | $0.00014 | $0.00218 |
| Haiku 4.5 | $0.00007 | $0.00109 |
Grade A, and why
exemplar-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 11d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Exemplar Audit
Find what the best implementations do that we don't, then close the gaps. This is the "measure twice" pass: compare our code against battle-tested references before declaring it done.
When to use
- After implementing a new subsystem (filesystem, protocol, cache)
- After a refactor that changed invariant-bearing code
- When reviewing code that implements a known standard
- When the user says
/exemplar-auditor/refine --pass 6 - When you suspect edge cases are missing but don't know which ones
Procedure
1. Identify exemplars
Check these sources in order:
- This monorepo first. The best reference for "how we do X" is often another subsystem that already does X well. GitHub Actions workflows, Terraform modules, Rust crates, Go controllers -- scan for the most mature sibling before looking externally.
- Module docs and READMEs in the subsystem being audited. Look for "References", "Inspired by", "See also", or citations to external projects.
storage.md,architecture.md, or equivalent design docs. These often name reference implementations.- The internet. Search for "canonical open-source implementation of {thing}". Prefer projects with >1000 stars, active maintenance, and a test suite.
Examples of good exemplars by domain:
| Domain | Internal exemplar | External exemplar |
|---|---|---|
| GitHub Actions | host-agent-rust.yml, host-agent-kvm.yml |
-- |
| Terraform | apps/cloud/apps/dcs/tf/main.tf |
checkov rules |
| Rust CI crate | host-agent (same workspace) |
-- |
| FUSE filesystem | -- | fuser SimpleFS, FUSE memfs |
| SQLite metadata | -- | JuiceFS pkg/meta/sql.go |
| WAL replication | -- | Litestream, libSQL bottomless |
| S3 chunk store | -- | JuiceFS pkg/object/s3.go |
| VFS layer | -- | Linux fs/fuse/dir.c |
| Content-addressed storage | -- | git sha1-file.c, Perkeep |
2. Clone or read
Clone into /tmp to avoid polluting the workspace:
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.
- 11d ago First seen · 117 lines · 70 tokens per session scan A adf934885cbd
exemplar-audit is a skill published in the GitHub repository Kevin-Liu-01/Agent-Machines (29 stars, last pushed yesterday), licensed MIT. It adds 70 tokens to every session and 1,092 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
dx-roaster
Brutally honest developer-experience audit for a GitHub repo. Scores 10 DX dimensions (time-to-first-success, README clarity, visual proof, install, quick-start, docs, examples, community, trust, marketing), writes a shareable roast in the requested tone (brutal/honest/kind), produces a prioritized action plan ranked…
mcp-review
Review MCP server specifications and updates for compliance, security, and quality. Use when evaluating server.json files, PRs adding/updating servers, or assessing MCP server changes. NOT for creating new entries (use add-mcp-server instead).
code-review
When the user wants a thorough code review of changes, a PR, a file, or a feature branch. Use when the user says "review this," "code review," "check my code," "what's wrong with this," "review my PR," "look over this," "critique this," "review before merge," or when submitting code for quality assessment. Performs…
holix-cron
Schedule recurring agent tasks via Holix built-in gateway cron (not crontab or custom scripts).
review-guidelines
Additional code review guidelines specific to this codebase. Auto-loaded by the review agent during PR reviews to enforce project conventions.
quick-review
Review a code diff or file for correctness bugs, security issues, and clear simplifications. One finding per line, severity-tagged. Use when user wants fast, focused code review.