agent eval for Claude Code

io.github.rudrendupaul/agent-eval

Statistical regression testing for LLM agents: p-value, effect size, and CI on behavior change.

client:Claude Code transport:stdio runtime:pypi

Install agent eval in Claude Code

run in your project directory

claude mcp add agent-eval -- uvx agent-regress-cli

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