Audit Session — Security skill for Claude Code
Find your own most-violated CLAUDE.md rules by auditing recent session transcripts.
How to install Audit Session
Installs to ~/.claude/skills/eivindkjosbakken-claude-code-setup-pack-audit-session/SKILL.md
mkdir -p ~/.claude/skills/eivindkjosbakken-claude-code-setup-pack-audit-session && curl -fsSL https://raw.githubusercontent.com/EivindKjosbakken/claude-code-setup-pack/HEAD/commands/audit-session.md -o ~/.claude/skills/eivindkjosbakken-claude-code-setup-pack-audit-session/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Audit Session does
description: Find your own most-violated CLAUDE.md rules by auditing recent session transcripts
/audit-session
Find which of my instructions the agent is actually failing to follow, ranked by frequency, so I mechanise the expensive ones instead of guessing.
What to do
**Locate recent transcripts.** They live under `~/.claude/projects//*.jsonl`, one file per session. Take the last 7 days unless I gave you a different window.
**Read my rules.** L
Alternatives in Security
- Memory Self Review — Mine recent agent history (claude-mem + usage stats) for recurring failures and repeated patterns, audit MEMOR 297 ★
- Unifly Audit — Security audit of UniFi configuration 248 ★
- Cross-Pollinate Hex MCP Servers — Diff-driven audit: find real transferable changes across hex-line-mcp, hex-ssh-mcp, hex-graph-mcp 238 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Token Waste Audit
Run the FR-26 token-waste audit — parse this project's Claude Code transcripts deterministically (requestId-de
Claudemd Sampling Audit
Retrospective batch scan of historical transcripts for 8 self-enforced HARD rules — text detectors (§10-V bann
Sessionxray
Security audit of Claude Code session transcripts: what did the agent actually touch.
Claude Desktop Tools
Native Windows 11 desktop application that acts as an autonomous visual command center to audit local storage,
Fenceline
Audit coding-agent session logs for boundary violations after the fact — forbidden commands, escapes outside a
Auditing Agent Config
Audit your Claude setup for drift: redundant skills, dead plugins, bloated CLAUDE.md, session-start pile-up.
Related Agents
Conversation Analyzer
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Triggere
Memory Extractor Agent
You extract perception changes from Claude Code session transcripts - the "aha moments" where understanding sh
Batch Session Reader
Reads multiple Fullstory session transcripts in parallel using isolated contexts. Use when you need to investi