Step 2 - Run Evaluation

Prerequisites

• Agent deployed and running in the selected environment
• Selected .foundry/agent-metadata*.yaml file loaded for the active agent root
• Evaluation suite selected from the environment's evaluationSuites[]
• For generated suites: suiteName present and verified with evaluation_suite_get
• For legacy suites: local dataset and evaluator metadata available in .foundry/

Definition of Done — Evaluation Run

A Step 2 evaluation run is complete only when every box below is checked. Do not produce a final "evaluation complete" summary, score table, or report link until all items are done. "Status reached completed" is not a stopping condition — evaluation_get returns metadata only.

• [ ] evaluation_agent_batch_eval_create returned an evalRunId
• [ ] evalId and evalRunId mirrored into the selected .foundry/agent-metadata*.yaml (environments.<env>.lastEval.\{evalId, evalRunId, runName, suiteName, suiteVersion, agentVersion, startedAt\}) so a later turn can resume
• [ ] Polling reached terminal state (completed, failed, or cancelled)
• [ ] Per-item output_items downloaded via the azure-ai-projects Python SDK (see Step 3 → Download Results) — NOT via evaluation_get, NOT via evaluation_dataset_sas_url_get