{ "artifact_type": "phase18-ci-release-reproducibility-pack", "artifacts": [ { "exists": true, "path": "docs/demo/phase16-blind-validation/regression-summary.json", "sha256": "c6063a141814d1f986c79f6b06c0a4fc81f4871583a36db27443ca974b55c8f6", "size_bytes": 1947 }, { "exists": true, "path": "docs/demo/phase16-blind-validation/route-diffs.jsonl", "sha256": "d787dd8b1a51964ef0370e4f59897011757b16b8e28cf146e799e44d84719333", "size_bytes": 16435 }, { "exists": true, "path": "docs/demo/phase17-calibrated-release-selector/release-decision.json", "sha256": "47c951630cae73c96f7ceae98c16ce7e15c263bac0b85abf4b9264bfcf24abdd", "size_bytes": 9679 }, { "exists": true, "path": "docs/demo/phase17-calibrated-release-selector/release-decision.md", "sha256": "90a76119c7c9ace581283d7366b7cf7a633585df70bc0aa39f092c6ec766dbf3", "size_bytes": 1908 }, { "exists": true, "path": "docs/demo/phase17-calibrated-release-selector/task-decisions.jsonl", "sha256": "164afe009addfb044f496bb0632f5e0fd1b3723719139d603dfaad5b60f45a2d", "size_bytes": 6026 }, { "exists": true, "path": "docs/demo/phase18-ci-release-reproducibility/release-check-summary.json", "sha256": "bad33f51a9b613740ad5d47c1a954165e34c5d98726a2ba11293e5c166084acf", "size_bytes": 903 } ], "commands": [ { "argv": [ "skilleval", "select-release-router", "--regression-summary", "docs/demo/phase16-blind-validation/regression-summary.json", "--route-diffs", "docs/demo/phase16-blind-validation/route-diffs.jsonl", "--output-dir", "docs/demo/phase17-calibrated-release-selector" ], "name": "select-release-router", "outputs": [ "docs/demo/phase17-calibrated-release-selector/release-decision.json", "docs/demo/phase17-calibrated-release-selector/release-decision.md", "docs/demo/phase17-calibrated-release-selector/task-decisions.jsonl" ] }, { "argv": [ "skilleval", "verify-release", "--public-root", "README.md", "--public-root", "docs/phase16.md", "--public-root", "docs/phase17.md", "--public-root", "docs/phase18.md", "--public-root", "docs/release-handoff.md", "--public-root", "docs/demo/phase16-blind-validation", "--public-root", "docs/demo/phase17-calibrated-release-selector", "--public-root", "docs/demo/phase18-ci-release-reproducibility", "--required-path", "docs/demo/phase16-blind-validation/regression-summary.json", "--required-path", "docs/demo/phase16-blind-validation/route-diffs.jsonl", "--required-path", "docs/demo/phase17-calibrated-release-selector/release-decision.json", "--required-path", "docs/demo/phase17-calibrated-release-selector/task-decisions.jsonl", "--required-path", "docs/demo/phase18-ci-release-reproducibility/release-manifest.json", "--required-path", "docs/demo/phase18-ci-release-reproducibility/release-manifest.md", "--required-path", "docs/phase17.md", "--required-path", "docs/phase18.md", "--required-path", "docs/release-handoff.md", "--summary-output", "docs/demo/phase18-ci-release-reproducibility/release-check-summary.json" ], "name": "verify-release", "outputs": [ "docs/demo/phase18-ci-release-reproducibility/release-check-summary.json" ] } ], "phase": "Phase 18", "reasons": [ "release decision, public release check, and artifact hashes are reproducible" ], "release_check": { "check_count": 5, "match_count": 0, "status": "PASS" }, "release_decision": { "approved_for_default": false, "baseline_router": "baseline-minilm", "candidate_router": "finetuned-embedding", "decision": "KEEP_BASELINE", "metric_deltas": { "mrr": -0.03125, "ndcg_at_5": -0.023067, "negative_accepted_rate": 0.0625, "negative_hit_rate": 0.0625, "recall_at_5": 0.0, "selection_rate_at_5": 0.0 }, "regression_count": 2, "selected_router": "baseline-minilm", "task_count": 16 }, "source_phase": "Phase 17", "status": "PASS" }