Search

Zero-trust execution firewall for autonomous AI agents (MCP/HTTP), with deterministic policy, approvals, and audit.

Detect and neutralize Unicode concealment codepoints in MCP tool metadata (covert-channel control). AGPL-3.0-or-later.

Validate OTA update packages end-to-end: signature chains, rollback protection, anti-downgrade counters, and delta-patch integrity.

Open-source governed memory for AI agents: prompt-injection-resistant writes, purpose-bound retrieval, provenance, and tamper-evident audit.

Live PoC: MCP attacks that compromise AI agents mid-session and how to block them in a few lines of code.

Python engine for Agent Threat Rules (ATR) -- the open detection standard for AI agents (like Sigma, but for prompt injection, tool poisoning, and MCP attacks). 311 rules, Cisco AI Defense shipped, 97.1% garak recall.

No description available

A practical, community-driven checklist for pentesting MCP servers. Covers traffic analysis, tool-call behavior, namespace abuse, auth flows, and remote server risks. Maintained by Appsecco and licensed for remixing.

AxioRank content-inspection engine: detectors + risk scoring for AI agent tool calls. Catches secrets, prompt injection, PII, destructive ops, and egress. Pure, universal, zero runtime dependencies.

Find what AI agents read but humans can't see — hidden prompt injection in files, web pages, PDFs, Office docs and MCP servers. CLI, Claude Code plugin, GitHub Action, library.

Scan any MCP server for prompt injection, tool poisoning, leaked secrets, and destructive capabilities. One command, no key, no signup.

Sandbox workspace tools: search, file read, DB queries, integrations. Returns synthetic data.

Static analysis tool for MCP repository security vulnerabilities

Security scanner for Claude MCP skills — scan before you install

Security audit framework for MCP (Model Context Protocol) servers

Security scanner for AI Agent skills, plugins, and MCP servers

The Model Context Protocol (MCP) represents a significant advancement in AI system integration, but with it comes a new landscape of security challenges. This repository provides a comprehensive guide to understanding, detecting, and preventing critical security vulnerabilities in MCP implementations.

Advanced SAST security analysis with multi-tool integration and comprehensive compliance verification

Fail-closed MCP tool baseline and runtime approval security for DeepSeek Harness.

Security check for AI-built projects, Claude Code skills and MCP servers