mcpsec
Active·★ 27·MIT·Updated 2026-08-13
mcpsec is a powerful security scanner and protocol fuzzer designed specifically for Model Context Protocol (MCP) servers, which connect AI agents to external tools. Unlike most MCP security tools that rely on static analysis, mcpsec connects to live servers to prove exploitation and offers extensive static audit capabilities with AI-powered taint analysis.
#Security Scanner#Protocol Fuzzer#MCP Security#Vulnerability Detection#Static Analysis#Runtime Analysis#AI-powered Security#DevSecOps
01
Features
01Runtime Security Scanning: Connects directly to live MCP servers via standard I/O or HTTP to identify vulnerabilities and map out attack surfaces in real-time.
02Advanced Static Code Auditing (Audit v3): Performs deep static analysis using a multi-stage pipeline, incorporating 3,450+ regex patterns, 149 Semgrep rules, AST-level analysis, and LLM-powered taint tracing for comprehensive code reviews.
03Protocol Fuzzing with AI Augmentation: Generates hundreds of high-intensity fuzz cases, leveraging AI for intelligent payload mutations, to expose protocol-level weaknesses, crashes, and denial-of-service vulnerabilities.
04Specialized Vulnerability Scanners: Includes a suite of dedicated scanners for specific vulnerability classes such as prompt injection, command injection, SQL injection, path traversal, SSRF, authentication bypasses, and dangerous tool chain detection.
05LLM Integration for Enhanced Analysis: Integrates with major AI providers (OpenAI, Anthropic, Google, etc.) to power features like advanced taint analysis, heuristic reachability, and sophisticated payload generation for more effective security testing.
02
Compatibility
Python
Runtime
Verified via docs
Nix
Package Manager
Verified via docs
MacOS
OS
Verified via docs
Ubuntu
OS
Verified via docs
Windows
OS
Verified via docs
OpenAI API
AI Provider
Verified via docs
03
Quick start
1
$ pip install mcpsec
04
Use cases
↳Continuous Security Assessment: Integrate into CI/CD pipelines to automatically scan local or remote MCP servers, ensuring ongoing security posture.
↳Pre-deployment Code Audits: Conduct thorough static analysis of MCP server source code from GitHub repositories or local paths before deployment, identifying critical vulnerabilities.
↳Vulnerability Discovery & Exploitation: Utilize fuzzing and specialized scanners to proactively discover new vulnerabilities and an interactive REPL for demonstrating exploitation.
↳Attack Surface Enumeration: Employ the `mcpsec info` command to map out the potential attack surface of an MCP server, understanding its exposed functionalities and risks.
↳Developer Security Tooling: Provide developers with a powerful tool to identify and fix security flaws early in the development lifecycle of MCP-based AI agent tools.
05
Alternatives
mcptrustchecker★ 172
Security scanner for MCP (Model Context Protocol) servers — reads the real published npm/PyPI source, not just metadata, to catch tool poisoning, prompt injection, toxic flows & supply-chain risk. Offline, deterministic A–F Trust Score, SARIF + CI gates.
cc-audit★ 26
AI-free static security scanner for Claude Code artifacts (Skills, Hooks, MCP configs). Detects data exfiltration, prompt injection, and supply chain risks with deterministic, reproducible results.
rea★ 358.1k
Reverse engineer anything with agents: one CLI and MCP server to inspect any app, understand it fully down to the binary level.
Reversecore_MCP★ 212
A security-first MCP server that empowers AI agents to perform automated reverse engineering, malware analysis, forensics, vulnerability research, and SAST — powered by Radare2, YARA, LIEF, Capstone, and more.
hackmyagent★ 83
Metasploit for AI agents: scan, attack, and fix AI agents and MCP servers. Open source security toolkit.
Comments
Log in to leave a comment
No comments yet. Be the first!