AgentTrust for Claude Code

io.github.vitaliiserbynassisterr/agenttrust

Quality verification for AI agents and MCP servers. 6-axis scoring, adversarial probes.

client:Claude Code transport:stdio runtime:pypi

Install AgentTrust in Claude Code

run in your project directory

claude mcp add agenttrust -e GROQ_API_KEY=<GROQ_API_KEY> -e CEREBRAS_API_KEY=<CEREBRAS_API_KEY> -e MONGODB_URL=<MONGODB_URL> -- uvx mcp-agenttrust

Adds it for this project only. Append --scope user to make it available everywhere. Claude Code docs