cc-steer audit

Audit a seeded stratified sample of the current prompt version’s verdicts.

cc-steer audit [OPTIONS]

The auditor is a stronger model, blind to the judge’s verdicts; its labels are keyed independently of the judge’s prompt version, so they accumulate across iterations and re-auditing a sampled row costs nothing. A pass that changes data syncs the dataset to HuggingFace; --no-sync skips it.

Full --help output
Usage: cc-steer audit [OPTIONS]

  Audit a seeded stratified sample of the current prompt version's verdicts.

  The auditor is a stronger model, blind to the judge's verdicts; its labels
  are keyed independently of the judge's prompt version, so they accumulate
  across iterations and re-auditing a sampled row costs nothing. A pass that
  changes data syncs the dataset to HuggingFace; ``--no-sync`` skips it.

Options:
  --accepts INTEGER             Audit budget for judge-accepted rows.
                                [default: 60]
  --rejects INTEGER             Audit budget for judge-rejected rows.
                                [default: 60]
  --seed INTEGER                Deterministic sampling seed (iteration
                                number).  [default: 1]
  --model [small|medium|large]  Auditor model tier.  [default: large]
  --concurrency INTEGER         Maximum concurrent claude subshells.
                                [default: 8]
  --db FILE                     Database path. Defaults to ~/.cc-
                                steer/feedback.db.
  --sync / --no-sync            Rebuild the derived dataset and push it to
                                your private HuggingFace repo when the pass
                                changed data.  [default: sync]
  --help                        Show this message and exit.

Options

--accepts: INTEGER = 60
Audit budget for judge-accepted rows.
--rejects: INTEGER = 60
Audit budget for judge-rejected rows.
--seed: INTEGER = 1
Deterministic sampling seed (iteration number).
--model: CHOICE = large
Auditor model tier.
--concurrency: INTEGER = 8
Maximum concurrent claude subshells.
--db: FILE
Database path. Defaults to ~/.cc-steer/feedback.db.
--sync, --no-sync
Rebuild the derived dataset and push it to your private HuggingFace repo when the pass changed data.