Quality Assurance

AGENTS.md · 6 documents

Maps every screen of an agent evaluation application to its URL, UI layout, and the evaluation features it implements.

title: 17. Evaluation Frameworks

Reports lint fixes and flake8 configuration for an NVDA addon, confirming 298 tests pass.

Defines the rules for a multiplayer party game where players shape AI bot personalities and compete to satisfy a secret judging criteria.

Defines a skeptical validation agent that tests completed work for correctness, completeness, and integration before accepting claims.

Defines a regression testing system for agents using golden datasets and hard/soft assertions to catch regressions after prompt changes.