Audit MCP Mock Drift — Security skill for Claude Code
Catch mock/live divergence in MCP eval golden tests (anti-pattern #2 from eval-mock-mode-discipline).
How to install Audit MCP Mock Drift
Installs to ~/.claude/skills/megabytespace-claude-skills-audit-mcp-mock-drift/SKILL.md
mkdir -p ~/.claude/skills/megabytespace-claude-skills-audit-mcp-mock-drift && curl -fsSL https://raw.githubusercontent.com/megabytespace/claude-skills/HEAD/commands/audit-mcp-mock-drift.md -o ~/.claude/skills/megabytespace-claude-skills-audit-mcp-mock-drift/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Audit MCP Mock Drift does
description: Catch mock/live divergence in MCP eval golden tests (anti-pattern #2 from eval-mock-mode-discipline) argument-hint: [--server=]
Audit MCP server eval golden tests for mock/live divergence: cases where `mock_response.body` claims a shape that the live API no longer returns. A stale mock silently teaches Claude wrong expec
Alternatives in Security
- Mcp-scan (Invariant Labs) — MCP security scanner with proxy mode for real-time scanning without infrastructure changes 1.9k ★
- Maestro Odyssey — Long-running iterative cycle — one entry, seven modes (debug improve planex review security defensive ui) 530 ★
- Agentseal — Security toolkit for AI agents 344 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Sovereign Skills
20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session man
Managed MCP Doctor
Audit Claude Code MCP governance drift, catch when allowedMcpServers no longer covers managedMcpServers (2.1.2
Ultra Test
Pre-delivery quality audit (Anti-Pattern + Coverage Gap + E2E + Performance + Security)
Review UX Audit
Full UX audit of a frontend codebase. Produces persona cards, journey maps, heuristic sweep, anti-pattern find
Draw
Run the anti-pattern selector standalone to verify the rotation rule and audit log.
Consistency Check
Audit a doc bundle for cross-reference integrity, count drift, pronoun ambiguity, audience assumptions, and ot
Related Agents
Eval Implementer
Implementer role of the D365FO agent eval loop. Runs ON THE VM (mcp-server in full mode + C# bridge) against t
Dashclaw Drift Auditor
Audits DashClaw's drift-prone hardcoded counts and version stamps — SDK method counts (Node/Python), MCP tool/
Test Authenticator
Detects fake, theatrical, and meaningless tests. Use PROACTIVELY after test suites are written to catch mock-e