Commit Graph

7 Commits

Author SHA1 Message Date
Aiden Bai 075865fc25 feat(evals): batch parity treatments in matrix (#1537)
* feat(evals): batch parity treatments in matrix

* fix(evals): close matrix evidence trust gaps

* fix(evals): bind matrix comparison evidence

* fix(parity): validate matrix artifact contents

* fix(parity): bundle matrix corpus evidence

* fix(parity): require lowercase matrix commits

* fix(evals): bound Daytona control-plane cleanup

* fix(evals): close matrix artifact review gaps

* fix(evals): align matrix verifier project hashing

* fix(evals): schedule matrix treatments together

* fix(evals): finalize matrix artifacts safely

* fix(evals): clean partial base publication

* fix(evals): preserve matrix retry batching
2026-07-31 06:44:08 -07:00
Aiden Bai f023fb9aab feat(evals): pair parity scans in one sandbox (#1536)
* feat(evals): pair parity scans in one sandbox

* test(evals): compare paired git directories portably

* test(evals): verify shared git directory semantically

* fix(evals): commit paired output atomically
2026-07-31 03:47:31 -07:00
Aiden Bai 7bbc7561fd feat(evals): add incremental parity caching (#1535)
* feat(evals): add incremental parity caching

* fix(parity): fail closed on uncertain scopes
2026-07-31 03:05:12 -07:00
Aiden Bai 2db2a97283 perf: reduce scan startup and workspace contention (#1533) 2026-07-30 19:42:52 -07:00
Aiden Bai 5b468f8a92 feat(rules): add React Three Fiber diagnostics (#1371) 2026-07-20 09:22:48 -07:00
Aiden Bai 3598138c7b Scale eval corpus and tighten diagnostics (#1403)
* scale eval corpus and tighten diagnostics

* fix review and Windows eval failures
2026-07-19 04:13:13 -07:00
Aiden Bai 7c9dbeda0d feat(evals): add Daytona corpus parity runner (#1384) 2026-07-19 00:21:46 -07:00