mcp-audit

MCTS (Model Context Threat Scanner) is a local-first security scanner for MCP servers -- static and live tool discovery, multiple analyzers, auditable risk scores, and JSON, SARIF, and HTML output. For authors and platform teams; CI-ready, no cloud API.

🛡️ Security audit CLI for Model Context Protocol (MCP) servers — scan AI agent configs for tool poisoning, rug pulls, hardcoded secrets, command injection & supply-chain risks. Pure Python, SARIF + CI ready.

Local-first TypeScript tooling for AI coding agents: MCP governance, multi-agent control room, and verification-as-a-tool.

Revenue triage CLI + public fit-check issue for products with attention but no paid conversion.

Agentic code review, security audit, and MCP-server auditing for Claude Code

SaSame MCP Factory — inspection/observation station: external readiness observations, signed MCP-Ready certificates, and curated portfolios of agent-callable tools.

MCP Infrastructure Audit Suite — Zero-install Claude Code skill. 30 governance rules, 39 protocol tests, health monitoring. Works with Claude Code, Cursor, Codex, Gemini CLI.

MCP gateway: a governance and audit reverse-proxy that sits in front of any MCP server. Per-tool RBAC, multi-tenant isolation, PII redaction on tool results, tamper-evident audit logging, per-tenant rate limits, HMAC request signing, all described in YAML.

Security scanner for MCP server configurations — npm audit for the AI agent era. npx mcp-audit-cli