| content |
{"Commits":[{"Sha1":"bfbc96902 {"Commits":[{"Sha1":"bfbc96902aa455160bff6085b7b60f936dfbe67b","Message":"docs(evidence): rebind DA05 post-D3 evidence to directory-pinned baseline 65d376a\n\nRe-ran P0-P3 against fixed baseline 65d376a (P0/P1 shadow-subset, selected\nsubset 453/453 and 1 failed/452; P2 fallback-full with graph accepted; P3 edge\nrewire GRAPH_EDGE_INVENTORY) and added P4: on a drift baseline swapping the\ncontracts and password-auth manifests, an identifier.test.ts M is rejected\nwith GRAPH_PACKAGES on a real Turbo graph. No full pnpm test comparison run.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-03T00:32:43-08:00"},{"Sha1":"65d376ab3b697ab6b5407fe277511ab948de2788","Message":"fix(da05): pin package directories so moved or swapped packages fail closed\n\nSecond independent review found package directories were only compared with the\nmanifests, never pinned (pre-existing since 02d2402): swapping the contracts and\npassword-auth directories, or moving contracts and leaving a stray\npackages/contracts/src/identifier.test.ts, kept names and edges, was accepted by\nreal Turbo, and the identifier rule then suggested @repo/contracts#test for a\nfile owned by another package or none. Pin the name-to-directory map measured on\n80c8e5a and add both drift negatives; sync CLAUDE.md / governance-experience.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-03T00:31:38-08:00"},{"Sha1":"44dbce11aa81f6d2a1804014b8e511295334e9fa","Message":"docs(evidence): rebind DA05 post-D3 evidence to exact-edge baseline dd42d82\n\nRe-ran all candidates against fixed baseline dd42d82: P0 benign and P1\nassertion-defect identifier.test.ts M suggest shadow-subset @repo/contracts#test\n(selected subset 453/453; 1 failed/452 passed); P2 amount+1 stays fallback-full\nwith the graph accepted; new P3 count-preserving edge rewire is rejected with\nGRAPH_EDGE_INVENTORY on a real Turbo graph. No full pnpm test comparison run.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-03T00:06:46-08:00"},{"Sha1":"dd42d826020e0a3c0860cf746ff29e7dadbb2a1c","Message":"fix(da05): pin exact post-D3 task edge list so count-preserving rewires fail closed\n\nIndependent review of 30386b1 found edges were still pinned by count only: moving\nweb's @repo/config dependency to api-nestjs, or contracts' public-file dependency\nto config (which changes the narrow closure), kept 14/13/10 and was accepted.\nPin the exact 13 dependent\u003edependency task edges measured on 80c8e5a plus task\nrows, and add two rewire negatives. Docs: state the exact-edge pin, scope the old\nGRAPH_PACKAGES fallback to D3 (56519ef) through 80c8e5a, and mark the 43b264f\ncomparison as historical.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-03T00:03:55-08:00"},{"Sha1":"c8a79121d017597682c6d69fc7e8c5a5fcefdb2e","Message":"docs(evidence): record DA05 shadow-subset on post-D3 pinned baseline 30386b1\n\nFixed baseline 30386b1 (selector re-pinned to the post-D3 graph). Clean\ncandidates: P0 benign identifier.test.ts M and P1 identifier assertion defect\nboth suggest shadow-subset @repo/contracts#test (selected subset run: 453/453\ngreen; 1 failed/452 passed); P2 amount+1 in quotation.ts stays fallback-full\nwith the graph accepted. No full pnpm test comparison was run this round;\nselection stays shadow-only and L1 is not enabled.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-02T23:46:02-08:00"}],"HeadCommit":{"Sha1":"bfbc96902aa455160bff6085b7b60f936dfbe67b","Message":"docs(evidence): rebind DA05 post-D3 evidence to directory-pinned baseline 65d376a\n\nRe-ran P0-P3 against fixed baseline 65d376a (P0/P1 shadow-subset, selected\nsubset 453/453 and 1 failed/452; P2 fallback-full with graph accepted; P3 edge\nrewire GRAPH_EDGE_INVENTORY) and added P4: on a drift baseline swapping the\ncontracts and password-auth manifests, an identifier.test.ts M is rejected\nwith GRAPH_PACKAGES on a real Turbo graph. No full pnpm test comparison run.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n","AuthorEmail":"hillao@juhailaoluodeMacBook-Pro.local","AuthorName":"juhailaoluo pro","CommitterEmail":"hillao@juhailaoluodeMacBook-Pro.local","CommitterName":"juhailaoluo pro","Timestamp":"2026-11-03T00:32:43-08:00"},"CompareURL":"luoanwu/juhai-quotation-system/compare/80c8e5a0dd4e458a4a5cbc9d6625ff61f4e4450e...bfbc96902aa455160bff6085b7b60f936dfbe67b","Len":6}... |