Marius CiepluchandClaude Opus 5 2dad25d312 fix(learning-harness): the round-4 review findings ...

Three defects, two of them the same shape as ones already rejected once:

- The meta-agent proposer still could not propose. `MetaAgent(inner_model=)`
  takes a ModelConfig; it was handed a model name, a `# type: ignore` hid
  that from mypy, and a bare `except Exception` turned the resulting
  AttributeError into `meta_agent_failed` -- which reads exactly like the
  model declining. The composition is now a seam a test can pin, and a
  construction error is re-raised rather than reported as a quiet round.
- `ousast pairs --hunter` crashed in its documented form: the K floor landed
  with `--k-runs` still defaulting to 1. The covering test always passed
  `--k-runs 4`, which is why nothing noticed.
- 28 rows (12% of the corpus) resolved a span on the vulnerable side and
  none on the fixed side, so every fixed-side finding read as outside the
  labeled function and the pair scored as silence whatever the detector
  said. `fix_function` now reaches the catalog and the scorer, and the
  generator derives it from the fixed excerpt for the 25 trap rows that
  never declared one. Measured after: 0 rows.

Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01T7XD7x5KonMdmGRkp63EQG

2026-09-06 14:03:10 +02:00

..

fixtures

feat(three-stage-scan): gate stage-2 maps against split-sink ranking

2026-09-04 11:53:06 +02:00

manifests

feat(three-stage-scan): add aliased-sink mutation the regex scan misses

2026-09-04 11:50:44 +02:00

measurements

test(learning-harness): the gate baseline and the degradation matrix

2026-09-06 13:20:23 +02:00

pairs

fix(learning-harness): the round-4 review findings

2026-09-06 14:03:10 +02:00