A local zero-trust checkpoint for the Model Context Protocol: read newline-delimited JSON-RPC, decide each message against a policy - allow, deny, redact - and write only what survives, with one audit line per message.
Cross-referenced across 55 tracked directories
#36273
Popularity Rank
1 / 55
Listed In
Emerging
Adoption Stage
12/16/2016
Created
159
GitHub Stars
Run an AI-powered security scan to analyze this package's source code for vulnerabilities, prompt injection vectors, data exfiltration risks, and behavior mismatches.
Scans fetch actual source code from the GitHub repository, not just the README.
Iman Schrock <[email protected]>
Trust & human sign-off for AI agents: approval required before irreversible agent actions
tumf
A secure shell command execution server implementing the Model Context Protocol (MCP)
sjkim1127
A security-first MCP server empowering AI agents to orchestrate Ghidra, Radare2, and YARA for automated reverse engineering.
...moreMaxFreedomPollard
Compartment - high-security, fully offline, encrypted vector memory for AI agents (MCP)
23
Forks
3
Open Issues
10/19/2026
Last Commit
Recently added to the ecosystem