mcp-safety is an open-source security scanner for Model Context Protocol servers. It helps developers inspect server capabilities and identify potential risks before installation or use in AI-agent workflows.
mcp-safety sits in PulseGate's AI & LLM security category. It focuses on assessing the security risks and capabilities of MCP servers before installing them. It is built as an open-source project for developers building or evaluating MCP-based AI agents. The project is open source (MIT). It ships for the command line, and it can be self-hosted.
mcp-safety first shipped in 2026. Key capabilities include MCP server scanning, static analysis, and capability inspection. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do