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 OKHP3/skillz --skill dependency-auditgit clone --depth 1 https://github.com/OKHP3/skillzWrote 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/okhp3/skillz/dependency-audit)<a href="https://agentmods.dev/skills/okhp3/skillz/dependency-audit"><img src="https://agentmods.dev/badge/skills/okhp3/skillz/dependency-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/okhp3/skillz/dependency-audit"><img src="https://agentmods.dev/badge/skills/okhp3/skillz/dependency-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 | $0.00083 | $0.01112 |
| Opus 5.5 | $0.00033 | $0.00445 |
| Sonnet 5.5 | $0.00017 | $0.00222 |
| Haiku 4.5 | $0.00008 | $0.00111 |
Grade A, and why
dependency-audit scanned grade A 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 1mo 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.
Unrestricted tool accesslowExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
- **Install scripts:** postinstall hooks that run arbitrary code at install time Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dependency audit
Most of the code shipping in a typical application was written by strangers. An audit answers three questions about that code: is it dangerous, is it legal, and is it maintained.
Scanners answer the first well. The second and third need judgement, and they are where the expensive surprises live — a licence discovered at acquisition, a critical library abandoned three years ago.
1. See the whole tree, not the manifest
Your manifest lists what you asked for. The lockfile lists what you got, and it is usually five to twenty times larger.
npm ls --all | wc -l # or: pip list, go list -m all, cargo tree
Transitive dependencies are where most risk sits, because nobody chose them. Note the depth and the total count — a project with 1,400 packages has a different risk profile from one with 40, whatever the direct list looks like.
Done when: you are working from the resolved tree, not the manifest.
2. Run the scanners, then triage the output
npm audit --json # pip-audit, govulncheck, cargo audit, osv-scanner
Scanner output is not a finding list; it is a starting list. Triage each:
- Is the vulnerable code path reachable from your usage? A deserialization CVE in a library
you only use for formatting is not exploitable.
govulncheckdoes this analysis properly; most tools do not. - Is it a dev dependency? A vulnerability in a build tool is real but has a different threat model than one in the server.
- Severity is not priority. A critical in an unreachable path ranks below a medium in your request-handling path.
Report reachability explicitly. A list of 200 unreviewed advisories teaches people to ignore the scanner, which is worse than not running it.
Done when: each advisory is triaged for reachability, not just listed.
3. Check licences before they become expensive
Licence problems surface at the worst moments — an acquisition, an enterprise sale, a distribution change, and are costly to fix retroactively.
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.
- 1mo ago First seen · 113 lines · 83 tokens per session scan A 7a0f09d85c95
dependency-audit is a skill published in the GitHub repository OKHP3/skillz (3 stars, last pushed today), licensed MIT. It adds 83 tokens to every session and 1,112 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
contract-review
A contract-review tool that checks agreements for basic, business, and legal risks. It adds comments without changing the original text and produces a review summary, overall opinion, and business process diagram.
law-to-markdown
A legal-document conversion skill for turning PDF, Word, or plain-text regulations into structured Markdown. Markdown is a text format that uses simple symbols for headings and lists.
antinet-four-color-cards
A document-analysis skill that organises findings into four card types: facts, explanations, risks, and actions. It is intended for financial reports and legal documents.
security-scan
A security and compliance check for files or URLs entering a system, producing a pass-or-reject decision and a report.
privacy-policy
Drafts a jurisdiction-aware privacy policy for any digital product. Use whenever a PMM or Product Manager needs to create, update, audit or review data protection documentation, asks about GDPR, CCPA or UK GDPR obligations, mentions "privacy policy", "cookie policy", "data retention", "right to be forgotten", "data…
kelly-legal-casebase-ingest
A review workspace for turning court judgments and arbitration awards into anonymized, structured internal case records. Court judgments and arbitral awards are written decisions that explain disputes and their outcomes.