{"branches":[{"author":{"email":"did:web:orbit-nodachi.test.pds.did.bot","name":"@orbit-nodachi.test.pds.did.bot","when":"2026-10-28T15:53:21.000000Z"},"committer":{"email":"noreply@tangled.sh","name":"Tangled","when":"2026-10-28T16:11:25.000000Z"},"message":"test(pull-numbers): read a stack expander's compact row off a real listing\n\nReplaces two hand-written compact-row tests with one excerpt of\nvibescrobble.com's merged listing, where #379 sits in #381's expander.\n\nCo-Authored-By: Claude Opus 5.5 \nChange-Id: Ie3787652430ca7923f77426723f306d266ca6524\n","name":"main","oid":"4d9b98114e1f7e2d9d7f0e7366a33d062e23a282"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-25T18:26:10.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-25T18:28:27.000000-04:00"},"message":"docs(readme): add the vibescrobble badge under the release line\n\nIt shows the agents live in this repository, as the didbot and\nvibescrobble.com READMEs do.\n\nCo-Authored-By: Claude Opus 5.5 \nChange-Id: I37c16a8fe15a33fa0e84527a1c09e25aee968605\n","name":"claude/readme-branding","oid":"e7a17f909b1125c9ebdf5c81b900149907e9280c"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-25T18:21:27.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-25T18:21:27.000000-04:00"},"message":"test(www): fail a build that loses its favicon\n\nverify-dist now checks that the root icons were built, that every page\nlinks them, and that the SVG favicon declares its namespace.\n\nCo-Authored-By: Claude Opus 5.5 \nChange-Id: I3480d91c1e3946dc0c14accb8568ac975b758f59\n","name":"claude/site-favicon","oid":"d62b8714cb3ce9c5be661400218ef6e6a76a249b"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:28:22.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-20T20:59:30.000000-04:00"},"message":"test(pr): prove --state all is a superset of --state open\n\nAdds an account with 600 pull records, five of them open at the\noldest end, past where the old five-page cap stopped reading. Also\nfixes the one existing assertion pinned to the old behavior: a\ncomplete `--state all` walk now says so, the same as `--state open`.\n\nChange-Id: Ib0f91c5442c49b913626ab76cfd6a9f8c1b18241\n","name":"claude/state-all-limit","oid":"1a886bcfbf300c6b205de0d4b40bc9401f75cd44"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-20T16:30:55.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-10-20T16:30:55.000000-04:00"},"message":"feat(plugin): add a Claude Code plugin that describes Tangled\n\nIts SessionStart hook says what Tangled is, how a didbot agent signs in to\natgc as itself, and to name the account on every command, then defers to\n`atgc agent`. It prints nothing when atgc is not on PATH.\n\nCo-Authored-By: Claude Opus 5.5 \nChange-Id: Ie0d639658cc44eab160725e0c8553004b1727713\n","name":"agents/atgc-plugin-development-setup","oid":"1037bf28771be91946fbf9a9356f57cf8d547529"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-16T21:56:49.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-16T21:56:49.000000-04:00"},"message":"fix(stack): prove the push per member, not over the whole range\n\nThe compare `stack create` used as proof answered about `target..branch`,\nand that answer grows with the stack: a twenty-commit branch came back as\n12.4 MB, past MAX_BODY, after the rewrite and the push. The knot is now\nasked for `target..` before anything is rewritten, and for each\nmember's own range after the push, so every answer is one change's size.\n\nChange-Id: I4416e29d73f7e8b9e799831670c08e8c880ed30b\n","name":"claude/stack-proof-size","oid":"8f2ce05c43c2c1829ac99285d70e938aad069478"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:12:36.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:13:19.000000-04:00"},"message":"docs(lexicon): fix renamed auth command in identity doc comment\n\nauth switch was renamed to auth default; this module's doc comment\nstill listed the old name.\n\nChange-Id: Ib0039540d4fc8a3bfa5a69bf35246d7add23e71f","name":"claude/doc-precedence-and-links","oid":"5dab84690d25eff5c013ade271b722650d167dad"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:04:33.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:08:39.000000-04:00"},"message":"docs(pr): fix stale file path in comparable_patch cross-reference\n\ncomparable_patch lives in src/cmd/stack/write.rs, not src/stack/write.rs;\nthe module moved under cmd/ and this doc comment was never updated.\n\nChange-Id: I48ebd056e71186e16d6e2a00ab1c1f9b94a4a551","name":"claude/three-doc-fixes","oid":"c77a603f256d7eaa4f39b540153562f22f893ad1"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:01:26.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T20:03:18.000000-04:00"},"message":"docs(style): remove reference to non-existent TODO.md file\n\nChange-Id: I26e4920f233089a36b67363a6b974a7ffdf39e0d","name":"claude/fix-doc-stack","oid":"b2b896d83a4dd10ec36ae4aa025eefee1fd7cf0f"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:56:51.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:57:31.000000-04:00"},"message":"docs(record): remove stale TODO.md reference from doc comment\n\nChange-Id: I22a2f5732253ebad344278356e3f68c86487f23b","name":"claude/doc-accuracy","oid":"69e446853936eb308a3e2c79011d0466d90542d8"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:56:08.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:56:48.000000-04:00"},"message":"docs(module-layout): fix cross-reference to environment docs\n\nREADME.md has no Environment section; the ATGC_* variable list lives\nin docs/environment.md.\n\nChange-Id: I9cd15ea09a24f0ebe708ef753dc4bf4e31bbceb4","name":"claude/fix-doc-inaccuracies","oid":"d99448b3f1d9af00fa1914885531d7e47a15d8e4"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:54:11.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:54:33.000000-04:00"},"message":"docs(pr): fix stale file path in comparable_patch cross-reference\n\nChange-Id: I1bed4e2183b328f41b0d8fb9887a328854ee8968","name":"claude/stale-docs","oid":"b450a21e71c4d07ffeae707f68840b61e3b970bc"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:42:09.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:42:09.000000-04:00"},"message":"test(env): pin the environment page to the variables the code reads\n\nAdds tests/environment_page.rs, which scans src/ and build.rs outside\ntest code for std::env::var/var_os, crate::env::switch and ATGC_*\nliterals, and fails naming any variable missing from\ndocs/environment.md. A short, commented allowlist covers names that\nare set rather than read, or that only tests use.\n\nTicks off the environment-page work in plan/docs.md against\n3mubapk23sf2s.\n\nChange-Id: Ia54bfac6ff1c1ca6a85d3eef9680191a93514974\n","name":"claude/env-page-complete","oid":"776ce634a560ae9fa4b4fcb303f241e1775bda4f"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:40:02.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:40:26.000000-04:00"},"message":"docs(testing): clarify that env::set_var is inherently unsafe\n\nChange-Id: Ie7a0d251fecea91fa7c18768cac01d718b9ba646","name":"claude/doc-fixes","oid":"4066716fae6eecbcdef8991a86e06654e8e09b05"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:29:52.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:33:04.000000-04:00"},"message":"docs(contributing): describe pr create's current default behavior\n\nf37703b made pr create push the branch by default and moved the old\npatch-alone behavior behind --patch-only; CONTRIBUTING.md still\ndescribed the old default.\n\nChange-Id: I59cd78f1b26b6261c7723e5308d65469e5445f01","name":"claude/doc-accuracy-fixes","oid":"6b5b2cd4b910a95674291cd8a57635eab84943ad"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:13:57.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:16:20.000000-04:00"},"message":"docs(cli): drop stale README pointer from completion help\n\nREADME.md no longer documents per-shell completion install steps;\nthe doc comment above already spells them out in full.\n\nChange-Id: I9642c12c12e3d9bf373065b9cf102af138556ed5","name":"claude/doc-fixes-stack","oid":"36c0ae1c5b5a03e4bdf5e9e610e9fc7dfe8fe110"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:05:26.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T19:06:37.000000-04:00"},"message":"docs(module-layout): fix outdated reference to README.md Environment section\n\nChange reference from nonexistent \"README.md's Environment section\" to\nthe correct location in docs/environment.md.\n\nCo-Authored-By: Claude Haiku 4.5 \nChange-Id: I9e6589c32210a2d16923b282b5e92d278f117160","name":"claude/docs-fixes","oid":"0540721109f5ca121fdf9bab24278083d080694a"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T18:53:22.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T18:53:22.000000-04:00"},"message":"docs: remove stale references to non-existent TODO.md\n\nThe TODO.md file was replaced by the plan/ directory structure. Update\ncomments in issue and stack modules to reference plan documentation instead\nof the removed TODO.md file.\n\nCo-Authored-By: Claude Haiku 4.5 \n","name":"claude/fix-stale-todo-refs","oid":"cfbcd034a6b4ff1dc96ca5cec8f7029c85dc76ae"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T15:20:27.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T15:20:27.000000-04:00"},"message":"fix(scope): weigh a repo scope's action list against a create\n\nA grant's `repo?collection=...` token can narrow itself with an `action`\nlist the same way it narrows itself by collection, and the coverage check\nignored it, so a session limited to `action=update` still passed as if it\ncould create records.\n\n`covers_create` reads that list the way the atproto OAuth scopes\nspecification defines it: absent is every action, and an explicit list that\nleaves `create` off is a real gap rather than a spelling to guess past.\n`repo_collections`'s doc comment now cites the specification for both\nparameters it reads past. Closes 3mudglso6bd22, \"atgc warns on its own\nrequested scopes\".\n\nCo-Authored-By: Claude Sonnet 5 \nChange-Id: I298a605b98887aeec7b4f956e39d9438a8e22dca\n","name":"claude/scope-collection-form","oid":"4cbdce685f06bea103d732f1dcd3de02e26df43e"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T15:12:21.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T15:12:21.000000-04:00"},"message":"feat(stack): count live and merged members on the stack view header\n\nMerged members stay in the listing because the bottom is the chain's base, so\na stack of three with one landed read like a three-member stack that was\nstuck. The header now says \"3 (2 live, 1 merged)\" once anything has merged,\nand is the line it always was when nothing has.\n\nChange-Id: I5c95b810bf2c2764ee063358e4b2ab8efb1954c2\n","name":"claude/lease-first","oid":"dab61568f6fd883f99e98a9ca1aee69ae19bfadf"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:35:21.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:35:21.000000-04:00"},"message":"fix(stack): keep the title a member's record already has\n\nA round no longer renames a pull to its bottom commit's subject, so the\ntitle `stack create` adopts and every `stack resubmit` passes over is the\none somebody wrote with `pr edit --title`. Only a member being opened for\nthe first time is titled by its commit.\n\nCo-Authored-By: Claude Fable 5.1 \nChange-Id: I5015af71138edc4fb2149bfb62c7acb7b0b394a0\n","name":"claude/keep-adopted-title","oid":"190df2ba5086e89739f1b8386c044b2a3445d20d"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:30:19.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:30:19.000000-04:00"},"message":"fix(repo): stop telling repo configure a declared hook is missing\n\ninstall_change_id_hook told a framework that already declares the\nchange-id script \"was not installed\" and to go add it -- the same\nlatent copy of 3mum4e67kbd2n flagged in the stack create fix. It now\nreads clients/git::hooks::declared too, via a new Installed::describe_given,\nand reports the hook as already in place instead.\n\nChange-Id: Ic8363c26e3b1735b6f10182f275e9326026267ff\n","name":"claude/hook-advice","oid":"cd63ccca9f2e1d270d3cc8d32804d94fe049beff"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:02:28.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T14:06:17.000000-04:00"},"message":"fix(exit)!: exit 429 as retryable, not as an unclassified failure\n\nA knot or PDS answering `429 Too Many Requests` fell through\n`exit::from_status` to the unclassified `1`, even though `clients/http.rs`\nalready treats 429 as transient and reads its `Retry-After`.\n\n`429` now joins `502..=504` at `6` (`Exit::Unreachable`), the code that\nmeans *retry unchanged*, so a caller such as `pr resubmit` that retries on\n`6` and gives up on `1` gets the right answer instead of stopping with the\nremote ahead of the last recorded round. Reported as `3murzx7klkd27`.\n\nCo-Authored-By: Claude Sonnet 5 \nChange-Id: Id6e2434fc5c4d56960b6516d2afc022566b9378e\n","name":"claude/exit-429","oid":"823344b275e92930d1c6037c60bc4ee22e3f91c2"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T12:22:29.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-09-05T12:22:29.000000-04:00"},"message":"docs(testing): keep the lesson the panel fixture taught\n\n`cmd::about`'s tests asserted the panel's shape against their own copy of it,\nwhich would have gone on passing after the real panel changed. main has since\ngiven `art` the one `panel_lines`, for a better reason than the copies had —\nthe terminal on atgc.codes is the command's own drawing rather than a picture\nof one — so the instance is gone and the rule was not written down anywhere.\n\nCo-Authored-By: Claude Opus 5 (1M context) \nChange-Id: I62fad65efb0490468e60cc8609f0456e2838e7db\n","name":"claude/home-newtype","oid":"394ad110d108196ba39474a4d581004f3103e686"},{"author":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-08-17T12:20:45.000000-04:00"},"committer":{"email":"did:plc:nlzmjyfv6loqtxyzvdcznwgf","name":"@permadeath.com","when":"2026-08-17T12:21:49.000000-04:00"},"message":"chore(deny): restore `wildcards = \"deny\"` now that nothing is a path dep\n\n`cargo deny check` passes as a gate again — `advisories ok, bans ok, licenses\nok, sources ok`. Two of the three findings from its first run on 2026-08-15\nwere one finding wearing two hats, and both went away with the lexicon crates:\nthe `[licenses.private]` block that excused them as unlicensed third-party code\nis deleted, and `wildcards` is back from `\"warn\"` to `\"deny\"`.\n\n`allow-wildcard-paths = true` goes with them. It only ever applied to a crate\nthat is itself private, which atgc is not, so it excused nothing — and it does\nnot reach vendor/jacquard-oauth either, whose `[patch.crates-io]` entry is not\na dependency of this manifest at all.\n\nTODO.md closes the publishability entry and rewrites the jacquard-oauth one to\nsay why it is now the only crates.io blocker left: cargo strips `[patch]` from\na published manifest, so a published atgc would resolve upstream jacquard-oauth\n0.12.1 and fail to compile on `lock_for_refresh`, a method only the fork has.\n`cargo publish --dry-run` verifies with the patch applied and cannot see it.\n","name":"claude/inline-generated-lexicons","oid":"c0b50956e48f621cd701471852e2f7c72bd011c7"}],"defaultBranch":"main","total":25}