A containment layer for AI agents that go rogue. Guard mode wraps any stdio MCP server: calls to tools you mark as risky wait for a person, the agent can't approve its own hold, and claimed accounts get Ed25519-signed records you verify offline. Covers the servers you wrap. MIT client.
Latest release v0.5.2 · 7 Oct 2026
1 file for Codex and OpenCode: vitna-compliance-mcp AGENTS.md — 716 tokens loaded in every session.
AGENTS.md C 716 tok These files are COSTRINITY/vitna-compliance-mcp's own configuration — they tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.