“agent-evals”
Agent Evals in CI: Suites That Run on Every Commit
Set up agent evaluation suites that run on every commit using the OpenAI Agents SDK, with GitHub Actions integration, regression detection, and PR commenting.
Trajectory Scoring: Rewarding Good Plans, Not Just Answers
Move beyond answer-level evaluation with trajectory scoring that rewards agents for good reasoning, efficient planning, and sound decision-making throughout their execution.
Tool-Selection Evals: Verifying the Agent Picks the Right Tool
Build eval suites that verify your agent selects the correct tool for each task, with precision metrics, confusion matrices, and regression testing for tool selection.
Multi-Turn Evals: Scoring Long Conversations End-to-End
Learn how to evaluate multi-turn agent conversations with trajectory scoring, context tracking, and end-to-end metrics that go beyond single-turn accuracy.
Evals for Enterprise Agents: From Unit Tests to Regression Suites
Build comprehensive evaluation suites for enterprise agents. Learn to create unit tests, regression tests, and benchmark suites for multi-step agent reasoning and tool selection.