security-audit SKILL.md skillA
Removed from GitHub on 2026-11-09. The file no longer exists in its repository, and every version below is still saved here. mdr add no longer finds it by name, but a lockfile that pins one of these versions still installs it from its recorded commit, for as long as GitHub keeps that commit.
security-audit from arcasilesgroup/ai-engineering is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add arcasilesgroup/ai-engineering/security-audit@sha256:fbb58fca9953452cPins the contents saved here, the only way to install this file now it is gone from GitHub. What a pinned install does.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- arcasilesgroup/ai-engineering
- Path
skills/ai-verify/tiers/3-chain/security-audit/SKILL.md- Versions recorded
- 2
- Latest version
- git:20260914.793714f, committed Sep 14, 2026
- First version recorded
- Sep 1, 2026
- Size
- 9,368 bytes
- Repository stars
- 65
- License
- Apache-2.0
Audit the codebase for exploitable security issues — traces attacker-controlled input to dangerous sinks across HTTP handlers, background jobs, auth and access control, and anything shipped to the client, and checks dependencies for known CVEs. Use when asked for a security review or audit, to check for vulnerabilities or leaked secrets, or before shipping code that handles auth, user input, or external requests.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Security audit Find issues an attacker could actually exploit. The unit of work is a **trace**: attacker-controlled input → the path it travels → the dangerous sink it reaches. A finding without that trace is a guess. Read `.claude/review/CONVENTIONS.md` first — stack detection, scope, severity, false-positive gate, report format, and the **installed-version rule**. That last one matters here more than anywhere: frameworks change their security defaults between versions, and half of what a reviewer "knows" about a framework's escaping, CSRF, or cookie behavior is version-specific. Verify against the installed source before reporting a missing protection the framework provides by default — or an assumed protection it does not provide. This is defensive review of code the user owns. Report vulnerabilities with enough detail to fix them — the trace and the fix, not a weaponized exploit. ## 1. Map the trust boundaries for *this* stack Before judging anything, write down which side of the wire each thing lives on. The boundaries differ by architecture, so derive them rather than assuming: - **Untrusted** — request bodies, query and path params, headers, cookies, form …
Read the whole file at its exact version.
How to install
The line at the top of this page is already the fingerprint pin, because a file GitHub no longer serves cannot be found by its label. With it the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_kb5ohco2wqkyu4py)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between Sep 1, 2026 and Sep 14, 2026, about one every 14 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260914.793714f latest | 2026-09-14 | 793714f | 9,368 B | A | view · diff |
| git:20260901.dca22e5 | 2026-09-01 | dca22e5 | 9,370 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (9368 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
arcasilesgroup/ai-engineering · 65 stars · license Apache-2.0 · pushed 2026-11-09 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_kb5ohco2wqkyu4py GET https://markdownregistry.com/api/v1/resolve?ref=arcasilesgroup/ai-engineering/security-audit GET https://markdownregistry.com/api/v1/blob/fbb58fca9953452c55eb74f41733d857bf6b0a6318ab97418413a6135a3dfc1f
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.