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 thixpin/pitway --skill security-auditgit clone --depth 1 https://github.com/thixpin/pitwayWrote 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/thixpin/pitway/security-audit)<a href="https://agentmods.dev/skills/thixpin/pitway/security-audit"><img src="https://agentmods.dev/badge/skills/thixpin/pitway/security-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/thixpin/pitway/security-audit"><img src="https://agentmods.dev/badge/skills/thixpin/pitway/security-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 | $0.00092 | $0.00753 |
| Opus 5.5 | $0.00037 | $0.00301 |
| Sonnet 5.5 | $0.00018 | $0.00151 |
| Haiku 4.5 | $0.00009 | $0.00075 |
Grade A, and why
security-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 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.
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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Security Audit
Fix causes, not symptoms; verify every trust boundary; leave a regression test behind.
Scope
Use for anything touching a trust boundary: security review, fixing a vulnerability, hardening a feature, or writing code that handles auth, permissions, secrets, or untrusted input.
Do not use for:
- General code quality with no security dimension — use
code-quality-review. - A quick automated pass over pending changes — the built-in
/security-reviewdoes that. This skill defines the standards to apply.
Root cause discipline
- For any vulnerability, identify the trust failure behind it (unvalidated input, missing authorization, confused deputy, secret in the wrong place) — then fix that, not just the reported instance.
- Search for the same pattern elsewhere in the codebase; vulnerabilities ship in families.
- Never "fix" a security finding by hiding the symptom (suppressing the error, filtering the payload string, disabling the scanner rule).
Authentication
- Every non-public endpoint/entry point verifies identity server-side; client-side checks are UX, never enforcement.
- Credentials: hashed with a modern KDF (bcrypt/argon2), never logged, never in URLs; sessions/tokens expire, rotate on privilege change, and invalidate on logout.
- Auth failures are uniform (no user-enumeration via differing errors/timing) and rate-limited.
Authorization
- Authorization is checked on every request at the resource level — object ownership, not just role ("can this user access this record", not "is this user logged in"). Missing object-level checks (IDOR) are the most common real-world hole.
- Deny by default; new endpoints and routes require an explicit permission decision.
- Enforcement lives server-side in one place (middleware/policy layer), not copy-pasted per handler.
Least privilege
- Code, services, tokens, and DB accounts get the minimum scope that works: read-only where read-only suffices, scoped tokens over master keys, short-lived over long-lived.
- New dependencies and integrations reviewed for what access they actually require.
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 · 59 lines · 92 tokens per session scan A 15d9b20baf20
security-audit is a skill published in the GitHub repository thixpin/pitway (20 stars, last pushed 2d ago), licensed MIT. It adds 92 tokens to every session and 753 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
review
Challenge a Hardproof implementation during REVIEW against its approved contract, code quality, tests, and risk boundaries.
behavior-implement
Implement one observable behavior or bug fix with red-green-refactor, writing a focused test that fails first, the minimum code that passes it, and then a behavior-preserving cleanup. Use for product behavior changes; for tooling or infrastructure, use it only when native checks such as linters, validators, dry runs…
cvg-work
Implement an agreed plan in behavior slices, validate the result, and check scope and unnecessary complexity before completion.
ai-slop-cleaner
Clean AI-generated code slop with a regression-safe, deletion-first workflow and optional reviewer-only mode.
tdd
Test-first at pre-agreed seams — one test, one implementation, tracer-bullet red/green; independent expected values; substitutions only at the seam's boundary class; launch Phase 4 reference.
adk-go-self-review
Review an ADK Go change the way a maintainer will — a fresh-context pass over the whole diff, five lenses (correctness and tests, scope, simplicity, style, adk-python parity), and the mutation check that proves your tests pin the change. Use before opening a PR, before any later push that changes code, and when asked…