Before you dive in
What you’ll find in the original
- Lint schemas before paid calls, revise against the tune split, and make regressions fail the comparison command.
- Run each example repeatedly to expose verdict changes and probability spread before choosing an automation threshold.
- Append question revisions and example hashes to a holdout ledger so threshold changes cannot silently reuse final evidence.
Worth knowing
The support-ticket figures are example-data results, not universal thresholds. The README also warns that short hashes can confirm guesses about enumerable private states.