Session Audit — Security skill for Claude Code
Self-audit pass over CLAUDE.md rules and memory files.
How to install Session Audit
Installs to ~/.claude/skills/benyoskovitz-claude-code-workflow-session-audit/SKILL.md
mkdir -p ~/.claude/skills/benyoskovitz-claude-code-workflow-session-audit && curl -fsSL https://raw.githubusercontent.com/benyoskovitz/claude-code-workflow/HEAD/commands/session-audit.md -o ~/.claude/skills/benyoskovitz-claude-code-workflow-session-audit/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Session Audit does
Self-audit pass over CLAUDE.md rules and memory files. Surfaces stale rules, undocumented workarounds, and contradictions between memory and current code. Read-only by default, never modifies CLAUDE.md or memory without explicit confirmation.
This pairs with `/session-end` (which it used to be part of) and with the workflow-meta lab: session-end is the deterministic wrap-up, session-audit keeps the ruleset honest on its own cadence.
**When to run:**
- After a long session where you referenced
Alternatives in Security
- Memory Self Review — Mine recent agent history (claude-mem + usage stats) for recurring failures and repeated patterns, audit MEMOR 297 ★
- Prompt Audit — Audit of Claude Code's default system prompt behavioral instructions 69 ★
- MCP Geo — Track brand citations across seven AI search surfaces 46 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Claudemd Rules
Audit the HARD-rules manifest. Cross-references spec/hard-rules.json with rule-hits.jsonl bySection data over
Audit Tokens
Analyze audit-log.jsonl for ghost-token-waste patterns across the 6 PLAN-047 detectors. Emits a markdown repor
Aiclean
Audit and clean up a Claude Code setup against Anthropic's current prompt-engineering guidance. Finds stale sc
Hub Audit
Monthly adversarial self-audit of the hub — a fresh, context-free worker hunts for drift, stale rules, and sel
Wiki Audit
Check wiki health — find broken links, stale content, contradictions, and auto-fix what it can
Agent Context Doctor
Audit agent context files like AGENTS.md, CLAUDE.md, .cursor/rules/.mdc, .github/copilot-instructions.md, and
Related Agents
Audit Drift
Detects SEO regressions by comparing current crawl against a stored baseline using 17 comparison rules at 3 se
Mb Developer
Generic memory-bank developer agent. Default implementer when no specialist role matches. Follows TDD discipli
Rule Validator
Rule Auditor. Use this agent to audit the Initial Roadmap against current rules/, draft missing rule files, an