Audit Prune Completeness — Security skill for Claude Code
Catch the github-mcp class of bug — CallTools handlers that exist but are invisible to Claude because ListTools never advertises them (orphaned), or ListTools entries that have no handler (zombies).
How to install Audit Prune Completeness
Installs to ~/.claude/skills/megabytespace-claude-skills-audit-prune-completeness/SKILL.md
mkdir -p ~/.claude/skills/megabytespace-claude-skills-audit-prune-completeness && curl -fsSL https://raw.githubusercontent.com/megabytespace/claude-skills/HEAD/commands/audit-prune-completeness.md -o ~/.claude/skills/megabytespace-claude-skills-audit-prune-completeness/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Audit Prune Completeness does
description: Catch the github-mcp class of bug — CallTools handlers that exist but are invisible to Claude because ListTools never advertises them (orphaned), or ListTools entries that have no handler (zombies) argument-hint: [--fix]
After every prune or forge, ListTools and CallTools can drift apart:
- Orphaned handler —
case/if
Alternatives in Security
- /web3 Audit — Smart contract security audit using the 10-bug-class methodology 927 ★
- Maestro Knowledge — Intent-driven knowledge-store and Run knowledge lifecycle management — audit/prune, stage candidates (with sig 530 ★
- Bountyforge — All-round bug bounty skill for Claude Code parallelized agents for smart contract audits (EVM, Move, Solana, T 399 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Audit MCP Error Semantics
Catch Resend-class bug (isError: false on HTTP 4xx/5xx) across all MCP server tool handlers
Extension Verify
Conformance and boundary audit of an existing cinatra extension repo: run the extension-authoring validate sta
Web3 Audit
Smart contract security audit — runs through 10 bug class checklist (accounting desync, access control, incomp
Skillhawk
Catch dangerous AI agent skills before they catch you. Zero-dependency security scanner for Agent Skills, SKIL
Audit MCP Mock Drift
Catch mock/live divergence in MCP eval golden tests (anti-pattern #2 from eval-mock-mode-discipline)
Managed MCP Doctor
Audit Claude Code MCP governance drift, catch when allowedMcpServers no longer covers managedMcpServers (2.1.2
Related Agents
Integration Enforcer
Verifies that new code is actually wired into the running system — not orphaned. Use PROACTIVELY after any imp
Wirer
Verifies that new code is actually wired into the running system — not orphaned. Use PROACTIVELY after any imp
Lead Triage
Judge ONE confirmation intent packet from lead-desk.py and return exactly three lines - class, would decide, b