mirror of
https://github.com/msitarzewski/agency-agents.git
synced 2026-10-22 11:14:52 +00:00
Adds promptfoo eval harness for agent quality scoring. LLM-as-judge system scoring task completion, instruction adherence, identity consistency, deliverable quality, and safety. Includes tests.