Review the judge's fix for quoting the worker's task as evidence, and close the gaps that review found #647

Merged
lilleman merged 9 commits from item-460 into main 2026-11-20 00:15:30 +01:00

9 Commits

Author SHA1 Message Date
Lilleman 6e22149334 Finish item 460 in the plan
CI / tests (pull_request) Successful in 4m32s
2026-11-20 00:10:43 +01:00
Lilleman ee296c332a Merge remote-tracking branch 'origin/main' into item-460 2026-11-20 00:10:42 +01:00
Lilleman a9bdef6fcb Merge remote-tracking branch 'origin/main' into item-460
CI / tests (pull_request) Has been cancelled
2026-11-20 00:06:58 +01:00
Lilleman 04c67d958e Answer stability round 2: 0 🔴, 0 🟡 (0 misread); tests and mutants clean at 3cce4043
CI / tests (pull_request) Successful in 3m23s
2026-11-20 00:02:03 +01:00
Lilleman afc51efb1b Answer stability round 1: 0 🔴, 0 🟡 (0 misread); decline the 🟢 on a NUL in the evidence, which the judge does not write 2026-11-20 00:01:03 +01:00
Lilleman 3cce40437e Let prepare-scoring read claude-config's scoring-run values from seat-prompts.md where docs/decisions.md holds none
CI / tests (pull_request) Successful in 3m53s
2026-11-19 23:39:54 +01:00
Lilleman 29b1be3fd7 Test how grounds_of splits an archived prompt, to kill the mutants that survived on its changed lines
CI / tests (pull_request) Successful in 3m49s
2026-11-19 23:25:38 +01:00
Lilleman 1845d8ff37 Answer prose round 1: 0 🔴, 4 🟡 (0 misread); say that the task and the transcript are searched, and where a quote may not run across them 2026-11-19 22:36:14 +01:00
Lilleman d954fb0026 Answer security round 1: 0 🔴, 0 🟡 (0 misread); keep a judge's evidence quote from running across the task and the tail, and name the task in the schema and the rejection message 2026-11-19 22:24:24 +01:00