mukul975/Anthropic-Cybersecurity-Skills
A comprehensive open-source library featuring 754 structured cybersecurity skills for AI agents, mapped to five industry frameworks to provide expert-level guidance.
Awesome MCP › Security & Reverse Engineering
radareorg/radare2-mcp
Radare2 MCP Server, or r2mcp, is an MCP server written in C on top of the native radare2 APIs that lets AI agents such as OpenCode, Mai, VS Code Copilot, Claude and CLION perform binary analysis and reverse engineering. It can run three ways: as a standalone command over stdin and stdout, as an r2 core plugin inside an existing radare2 session, and as an HTTP MCP server. In plugin mode the r2mcp command reuses the current RCore, so tools operate on the file and analysis state already loaded in that session, with subcommands to start, stop, restart and check the server, control debug logs, manage supervisor approvals, and list the r2mcp.* eval keys that mirror the CLI options. HTTP mode is enabled with -H and a port, binding to 127.0.0.1 unless 0.0.0.0 is given explicitly; -X enables per-session state routed by the X-Session-ID header, and -a sets a bearer token that clients must send, with -A or -a random generating one at startup and printing it to stderr. Per-session state and bearer auth need radare2 ABI 91 or newer. Security controls include a read-only mode, a sandbox lock with adjustable granularity, and fine-grained enabling or disabling of individual tools, while raw access to run r2 commands or r2js scripts and other dangerous tools stay behind approvals unless yolo mode is turned on. It can also attach to any local or remote r2 or iaito session through r2pipe. Installation is easiest with r2pm -Uci r2mcp, and Docker and make install builds plus a Codex plugin target are documented, alongside client configuration for Claude Desktop, VS Code and Zed.
https://github.com/radareorg/radare2-mcp
Also in Security & Reverse Engineering
NVIDIA/SkillSpector
NVIDIA security scanner for AI agent skills and MCP tooling that detects prompt injection, data exfiltration, MCP tool poisoning and least-privilege problems, and can itself be run as an MCP server inside agent sessions.
mrexodia/ida-pro-mcp
ida-pro-mcp is an MCP server for IDA Pro that enables advanced reverse engineering capabilities through MCP-based interactions and automation.
LaurieWired/GhidraMCP
GhidraMCP is an MCP server that integrates Ghidra's reverse engineering capabilities with MCP clients, enabling automated binary analysis and decompilation through large language models.
invariantlabs-ai/mcp-scan
MCP-Scan is a security tool that statically and dynamically scans and monitors Model Context Protocol (MCP) connections to detect and prevent vulnerabilities such as prompt injections, tool poisoning, and cross-origin escalations.
zinja-coder/jadx-ai-mcp
JADX-AI-MCP is a JADX plugin integrating Model Context Protocol to enable AI-powered live reverse engineering, vulnerability detection, and code analysis of Android APKs using large language models like Claude.
mariocandela/beelzebub
Beelzebub is a secure low-code honeypot framework leveraging large language models and the Model Context Protocol (MCP) to detect and analyze cyber attacks, including prompt injection attempts against LLM agents.
svnscha/mcp-windbg
mcp-windbg is a Model Context Protocol server that enables AI models to analyze Windows crash dumps using WinDBG through natural language interaction and command execution.