eval
ak eval ui-benchmark
Measure independently labeled UI signals offline
Use ak eval ui-benchmark for the operation described below. The command help records its arguments, effects, and options.
Command contract
Measure independently labeled UI signals offline
Synopsis
Replays bound typed responses against independent assertion labels. Synthetic fixtures validate integration, not production quality or savings. Actual provider, reasoning and vision calls are zero; recorded measurements require provenance.
Effects: Reads local records; no network calls or writes.
Exit status: 0 valid benchmark; inspect useful, missed, and unknown counts 1 invalid corpus or replay
ak eval ui-benchmark <corpus.json|-> [flags]Examples
ak eval ui-benchmark corpus.json --json
ak eval ui-benchmark - --jsonOptions
-h, --help help for ui-benchmark
--json Emit machine-readable JSON (implies --no-interactive)
--no-interactive Disable interactive prompts (CI-safe)
-q, --quiet Suppress non-error output on stderr
-V, --verbose Extra diagnostic output on stderr (loses to --quiet)
-y, --yes Assume yes for all promptsSEE ALSO
- ak eval - Run controlled skill and agent evaluations