lh-code-audit SKILL.md skillA
lh-code-audit from jrodrigopuca/local-agents is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add jrodrigopuca/local-agents/lh-code-audit@v1.1Pins v1.1 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- jrodrigopuca/local-agents
- Path
security/skills/lh-code-audit/SKILL.md- Versions recorded
- 1
- Latest version
- v1.1, committed Oct 14, 2026
- Size
- 5,714 bytes
- Repository stars
- 2
- License
- Apache-2.0
Exhaustive code auditing for vulnerability classes: injection, authz, auth, secrets, crypto, deps, and the method for reading code adversarially. Trigger: load when auditing code for security flaws, reviewing a diff for vulnerabilities, or hunting a specific vuln class.
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
## Critical Patterns ### 1. Read like an attacker: follow tainted data, not the control flow Don't read the code the way it runs — read the way data ENTERS and chase it. Trace every input from its source (the door) to its sinks (query, command, HTML, filesystem, response). A vulnerability is tainted data reaching a dangerous sink without a sanitizing gate in between. `rg` the sinks first (query builders, `exec`, `innerHTML`/`dangerouslySetInnerHTML`, `eval`, file ops, deserializers), then walk backwards to see if attacker-controlled data can reach them. Where an LLM sits in the flow, its output is tainted by everything it read, so treat it as a SOURCE, and treat every tool or query the model can invoke as a sink: the audit question is whether content the model ingested can steer a call the user never made ([data-ml/lh-llm-integration](../../../data-ml/skills/lh-llm-integration/SKILL.md) bounds the call; you audit what it can reach). ### 2. The high-yield vulnerability classes (audit in this order of ROI) | Class | The question | Where it hides | |-------|--------------|----------------| …
Read the whole file at its exact version.
How to install
mdr add jrodrigopuca/local-agents/lh-code-audit@sha256:1a198ecd69c267f3The line at the top of this page pins the label v1.1, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way 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_7xxz7z67uxzl2yzg)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 14, 2026.
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 (5714 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
jrodrigopuca/local-agents · 2 stars · license Apache-2.0 · pushed 2026-10-23 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_7xxz7z67uxzl2yzg GET https://markdownregistry.com/api/v1/resolve?ref=jrodrigopuca/local-agents/lh-code-audit GET https://markdownregistry.com/api/v1/blob/1a198ecd69c267f37f9cf8d88728b4de85581547f3e9333ec0e7b6de6c18d71b
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.