OptionalagentThe target agent's default Evaluation rubric, used when nothing earlier names one.
OptionalagentAgent/system version being tested
OptionalconfigOverride test configuration
OptionaldryValidate configuration without executing
OptionalenvironmentEnvironment context (dev, staging, prod)
OptionalgitGit commit SHA for versioning
OptionallogLog callback for streaming execution details to the test run log
OptionalprogressProgress callback for real-time updates
OptionalpublishedLatest published version observed at suite-run start, used to pin the suite.
OptionalrepeatOverride the test's RepeatCount field at runtime. Useful for flaky-test detection: forces every test in the run to execute N times so the suite runner can compute score variance. If undefined, the test's own RepeatCount is used.
OptionalrubricExperiment override for the rubric that judges this run. A version here is explicit and is not replaced by the suite pin.
OptionalrubricExplicit rubric version id. When omitted, the suite run pins the latest published version.
OptionaltagsTags to apply to the test run (JSON string array)
OptionalvariablesVariable values to use for this run. These values take highest priority in the resolution order: run > suite > test > type
OptionalverboseVerbose logging
Options for running a single test