Claudemd Bypass Audit — Security skill for Claude Code
R3 Step 2 — lesson-bypass detector.
How to install Claudemd Bypass Audit
Installs to ~/.claude/skills/sdsrss-claudemd-claudemd-bypass-audit/SKILL.md
mkdir -p ~/.claude/skills/sdsrss-claudemd-claudemd-bypass-audit && curl -fsSL https://raw.githubusercontent.com/sdsrss/claudemd/HEAD/commands/claudemd-bypass-audit.md -o ~/.claude/skills/sdsrss-claudemd-claudemd-bypass-audit/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Claudemd Bypass Audit does
name: claudemd-bypass-audit description: R3 Step 2 — lesson-bypass detector. Joins memory-prompt-hint `suggest` events with subsequent transcript activity to compute cite-recall (applied / (applied + bypassed)) across recent sessions. Makes the §11 MEMORY.md read-the-file effectiveness observable.
Default window is 30 days. The script answers one question: when claudemd's `memory-prompt-hint` hook surfaced a relevant memory to the agent, did the agent actually read or cite it?
**$ARGS
Alternatives in Security
- Security Ownership Map — Map people-to-file ownership, compute bus factor, and identify risks 14.6k ★
- Token Scan — Meme coin and token security scan — checks for rug pull vectors (hidden mint, honeypot, fee manipulation, LP l 3.8k ★
- Loongsuite Pilot — Local-first telemetry collector for AI coding agents — unified OpenTelemetry events for Claude Code, Codex, Cu 150 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Claudemd Audit
Aggregate claudemd rule-hits over the last N days. Shows top banned patterns, hook deny counts, per-spec-secti
Courseware Gen
Generate the WSQ courseware for this course — the slide deck (PPT), Lesson Plan (LP) and Learner Guide (LG) pl
Claude Bash Tutor Skill
Claude Skill teaching bash from the ground up: 17 units, prompt basics through security and debugging. Every e
Vigilant Guacamole
AI Code Review Agent (with Auto-Fix) Connects to GitHub PRs, runs parallel Security / Performance / Test-cover
Claude Rank
SEO/GEO/AEO audit plugin — tells you why AI won't cite your site, then auto-fixes robots.txt, sitemap.xml, llm
Aegis Run Logger
Stop hook that archives every Claude Code session transcript to .aegis/brain/runs/ for later replay. Use this
Related Agents
Lesson Cut Critic
Reads the simulated delivered transcript of a lesson after every proposed cut is applied, and finds what per-w
User Sim
Plays a researcher or applied-AI engineer using whileai for real - their own keys, their own compute (Modal, P
LLM Checker
LLM training-bias pattern detector — finds structural patterns from tutorial/library training data applied to