Commit Graph
5 Commits
Author SHA1 Message Date
Marius CiepluchandClaude Opus 5.5 edcf145b00 spec(learned-decision-engine): tasks approved; tasks 1-4 first, harvest waits for budget
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
2026-09-30 18:14:53 +02:00
Marius CiepluchandClaude Opus 5.5 d7a572498d spec(learned-decision-engine): design approved; implementation plan for review
Assumed-benign pushes mined from non-security commits as a separate label source
(maintainer decision). Ten tasks in the design's commit order; the harvest,
exp-001 and the v3 check wait for budget go-ahead.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
2026-09-30 18:14:31 +02:00
Marius CiepluchandClaude Opus 5.5 f6f9591cd0 spec(learned-decision-engine): design for review
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
2026-09-30 18:13:27 +02:00
Marius CiepluchandClaude Opus 5.5 0fad93fcc5 spec(learned-decision-engine): requirements approved
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
2026-09-30 17:59:40 +02:00
Marius CiepluchandClaude Opus 5.5 afde23d3cf spec(learned-decision-engine): brief and requirements for review
Maintainer direction: A/B testing to extrapolate detection instead of hard-coding
it, and a trained decision engine. Instruments become signal producers; labels
from ground truth with provenance; a simple calibrated engine with population
folds; experiments as plane Runs; extrapolation with intervals, checked once on
frozen v3; the engine is what scan/pre-push/plane report.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
2026-09-30 17:59:06 +02:00