The 15 most popular security MCP servers out of 260, with a security level for each.
| # | Name | Popularity | Security |
|---|---|---|---|
| 1 | semgrep Scan code for security vulnerabilities using Semgrep. | 17.1k ★ | Low risk |
| 2 | IDA Pro IDA Pro plugin for binary analysis: decompilation, disassembly and automatic malware analysis reports. | 13k ★ | Low risk |
| 3 | Ghidra Autonomous reverse engineering with Ghidra: decompile binaries, rename methods and data, and list methods, cla | 11.4k ★ | Low risk |
| 4 | jadx-ai-mcp JADX decompiler plugin providing live, LLM-assisted reverse engineering. | 2.9k ★ | Low risk |
| 5 | beelzebub Honeypot framework for building decoy MCP tools that an agent would never use in normal work, to detect prompt | 2.2k ★ | Low risk |
| 6 | vet Locally check npm and PyPI packages, such as those suggested by AI coding tools, for vulnerabilities and malic | 1.1k ★ | Low risk |
| 7 | skylos Dead code detection, security scanning and code quality analysis for Python, TypeScript and Go, with AI-assist | 848 ★ | Low risk |
| 8 | GhidrAssistMCP Native Ghidra integration with GUI configuration, logging and no external dependencies. | 776 ★ | Low risk |
| 9 | apktool-mcp-server Automate reverse engineering of Android APKs with Apktool. | 674 ★ | Low risk |
| 10 | emilia-protocol Requires a named human's approval before irreversible agent actions such as payments or deploys, then issues a | 611 ★ | Low risk |
| 11 | binary_ninja_mcp Binary Ninja plugin and bridge for automating binary analysis and reverse engineering. | 455 ★ | Low risk |
| 12 | radare2-mcp Disassemble and inspect binaries for reverse engineering with the Radare2 disassembler. | 328 ★ | Low risk |
| 13 | DashClaw Fail-closed approval layer for unattended agent runs: checks actions against org policy, requests human approv | 313 ★ | Review before installing |
| 14 | Wireshark-MCP Wireshark packet analysis: capture, protocol statistics, field extraction and security analysis. | 308 ★ | Low risk |
| 15 | OpenWorkProof Verifiable execution protocol for agent tool calls: signed policy decisions, causal evidence chains and offlin | 287 ★ | Review before installing |