llm-ingestion-okf/tools
Kjell Tore Guttormsen bc0b4130f1 feat(tools): okf_contract_check reads the contract's checkable half
Fourteen rules, each with its own code: one "invalid" verdict over fourteen
different defects is a diagnostic no caller can act on. The report quotes its
own denominators -- rules run, excerpts and withheld entries examined -- because
a checker that exempted itself from section 5 would be stating the rule it
breaks.

Three exit codes, not two. "The check did not run" and "the check failed" are
different outcomes, and an unread file reported as a failed check is the fourth
face of the verification law.

The shipped template plus its example payload is the known-positive arm, so a
checker that refuses everything cannot be green on the thirteen negative ones.
A last test asserts every literal the checker enforces appears in the contract
document: two copies of a closed set drift, and the copy nobody reads is the
one that goes wrong.

996 -> 1023 tests.
2026-09-02 16:09:27 +02:00
..
okf_adjudicate.py feat(tools): adjudication command with a switchable model leg 2026-09-02 14:54:20 +02:00
okf_contract_check.py feat(tools): okf_contract_check reads the contract's checkable half 2026-09-02 16:09:27 +02:00
okf_corpus_run.py feat(tools): corpus harness reporting K1 and K2 with denominators 2026-09-02 15:40:06 +02:00
okf_fidelity.py feat(tools): re-measurable K2 fidelity instrument with a negative control 2026-09-02 15:08:08 +02:00
okf_propose_segments.py feat(segmentation): quote anchors so a re-extraction costs a re-anchor 2026-09-02 14:42:16 +02:00
okf_watch.py feat(watch): weekly OKF upstream watch that can prove it found nothing 2026-08-23 20:38:37 +02:00