If you discover a security vulnerability in MCP Guard itself, please report it responsibly:
- Do NOT open a public GitHub issue
- Email the maintainers or use GitHub's private vulnerability reporting feature
- Include steps to reproduce and potential impact
We will respond within 48 hours and work with you on a fix.
MCP Guard is a static analysis tool. It reads files but does not execute MCP servers or connect to external services. The attack surface is limited to:
- Maliciously crafted source files that could exploit the Python AST parser
- Maliciously crafted JSON config files
| Version | Supported |
|---|---|
| 0.1.x | Yes |