Expand description
Runs and case runs: the durable side of an evaluation.
Structs§
- Cancel
RunRequest - Case
RunPage - Case
RunSummary - One planned execution of one case.
snapshot_refis frozen at run creation, so a later binding edit cannot retroactively change it. - Create
RunRequest - Eval
Operation - The durable evaluate task. Its id is the
evaluateTaskIda driver uses. - Evaluation
Subject - Agent id is an optimistic precondition; the Runtime remains the authority on the agent/runtime binding and validates it at execution time.
- Experiment
Variant - GetOperation
Request - List
Case Runs Request - Snapshot
Selection
Enums§
- Case
RunState - Terminal classification of a case run. The five outcomes are deliberately distinct: a flaky sandbox, a broken case and a genuine quality miss demand completely different responses.
- Operation
State