coval-ai/Coval Run Evaluation

Launch and monitor AI agent evaluation runs using the Coval v1 Runs API

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
agent_idThe agent ID to test (22 character identifier)yes
persona_idThe simulated persona ID to use for testing (22 character identifier)yes
test_set_idThe test set ID containing test cases to run (8 character identifier)yes
metric_idsOptional JSON array of metric IDs to evaluate (e.g., ["metricId1", "metricId2"]). If not provided, uses agent's default metrics.no
iteration_countNumber of times to run each test case (1-10)no1
concurrencyNumber of simulations to run concurrently (1-5)no1
metadataOptional JSON object for custom metadata (e.g., {"campaign_id": "q4_2025", "created_by": "ci/cd"})no
max_wait_timeMaximum time to wait for run completion in secondsno600
check_intervalInterval between status checks in secondsno30
api_base_urlBase URL for the Coval APInohttps://api.coval.dev/v1
fail_on_metric_idMetric ID to check for failure conditions. If any simulation has the fail_on_metric_value for this metric, the action will exit with code 1.no
fail_on_metric_valueValue that indicates failure for the specified metric (e.g., 'NO'). Case-insensitive comparison.no
namedescription
run_idThe unique identifier of the launched run
statusFinal status of the run (COMPLETED, FAILED, etc.)
run_urlURL to view the run in Coval dashboard
metric_failuresJSON array of simulation IDs that failed the metric check (empty if none)