Commit Graph

  • 6e05fdc209 docs: Cold-Start-Prompt fuer Evaluate/Review/Release ergaenzt chore/scrub-private-ip Tarik Moussa 2026-07-31 01:34:11 +02:00
  • 15b782fa8d Merge pull request 'chore: interne Host-Adresse durch generischen Hostnamen ersetzt' (#17) from chore/scrub-private-ip into main main user2595 2026-07-30 22:52:58 +00:00
  • 53c40f6da2 chore: interne Host-Adresse durch generischen Hostnamen ersetzt Tarik Moussa 2026-07-31 00:51:27 +02:00
  • 5aa2f7db40 Merge PR #16: data-quality roadmap R-A–R-F + DQ-5 (reconciled, reviewed) user2595 2026-06-27 07:12:29 +00:00
  • ab621fc5f5 test(review): cover findings #1 (non-bare caller id) and #4 (backfill connection scope) Tarik Moussa 2026-06-27 09:09:39 +02:00
  • b2f01ce010 fix(review): address PR #16 code-review findings Tarik Moussa 2026-06-26 22:50:23 +02:00
  • 9e5184385a merge: reconcile data-quality roadmap with audit-remediation main (#12-15) Tarik Moussa 2026-06-22 15:11:26 +02:00
  • e817022097 docs: mark integration/roadmap merge-ready (391 green) + cleanup notes Tarik Moussa 2026-06-22 02:01:38 +02:00
  • 2d78f9c48e merge: infra TODO — tmux-on-Jetson SSH stability Tarik Moussa 2026-06-22 01:59:48 +02:00
  • 6911f0065e merge: R-D (citing-coverage warning in graph report) Tarik Moussa 2026-06-22 01:59:48 +02:00
  • aaa86dee2f merge: R-B (Crossref references as third citation fallback) Tarik Moussa 2026-06-22 01:59:02 +02:00
  • 42608c73b8 merge: R-C (.tex ingest) + R-F (section labels) + DQ-5 arXiv fix Tarik Moussa 2026-06-22 01:58:43 +02:00
  • 49c251eafa docs: open-items — Jetson back (tunnel had dropped), R-F accent re-clean done Tarik Moussa 2026-06-22 01:54:32 +02:00
  • f761ee63f8 docs(audit): R-F accent re-clean done — 8 labels fixed in place, 0 artifacts remain Tarik Moussa 2026-06-22 01:53:36 +02:00
  • 200b7f8188 docs: session 2026-06-17 open-items handoff notice Tarik Moussa 2026-06-22 01:48:47 +02:00
  • a050f6622a docs(infra): TODO — tmux-on-Jetson + autossh for stable long-running ops Tarik Moussa 2026-06-18 09:32:03 +02:00
  • 5c60a7eda4 docs(audit): add R-F (richer section labels) resolution — 10% to 95% non-body Tarik Moussa 2026-06-18 09:25:03 +02:00
  • 1a879d4827 feat(quality): clean LaTeX accents/ties + word-boundary truncation in section titles Tarik Moussa 2026-06-18 09:17:43 +02:00
  • c70ff9aa1e feat(quality): store real section titles for .tex chunks (R-F) Tarik Moussa 2026-06-18 09:03:46 +02:00
  • 5672358bbe feat(graph): warn on low citing-paper coverage in graph report (R-D) Tarik Moussa 2026-06-18 07:59:25 +02:00
  • c22c0db3d0 docs(audit): mark R-C done (section-aware .tex ingest + DQ-5 arXiv half) Tarik Moussa 2026-06-18 07:47:43 +02:00
  • f504ca62db fix(openalex): canonicalize arXiv DOI to bare arXiv id (DQ-5 arXiv half) Tarik Moussa 2026-06-18 04:25:22 +02:00
  • 121b582f46 fix(arxiv): handle legacy single-file gzipped source (no tar wrapper) Tarik Moussa 2026-06-18 04:18:18 +02:00
  • 1da81c7b28 feat(tex): section-aware multi-file .tex ingest (R-C) Tarik Moussa 2026-06-18 04:15:20 +02:00
  • bf90c6cd57 feat(grobid): run GROBID on-demand (restart "no") to free ~3.5GB when idle Tarik Moussa 2026-06-17 21:36:04 +02:00
  • 82cea2a33b feat(crossref): reference extraction as third citation fallback (R-B) Tarik Moussa 2026-06-17 21:32:46 +02:00
  • 1516684bbb feat(grobid): R-A reference backfill + native arm64 GROBID on Jetson Tarik Moussa 2026-06-17 08:28:34 +02:00
  • b55aa5b429 chore: gitignore generated wiki/index.md Tarik Moussa 2026-06-17 00:42:37 +02:00
  • 0021859094 fix(ingest): normalize GROBID-derived citation DOIs to bare lower-case Tarik Moussa 2026-06-17 00:39:04 +02:00
  • b371119264 docs(audit): scope DQ-3 to txt→chunks + open DQ-6 (PDF→txt extraction fidelity) Tarik Moussa 2026-06-17 00:33:45 +02:00
  • ea73b1475c fix(openalex): normalize DOI to bare canonical id so re-ingest is idempotent (DQ-5) Tarik Moussa 2026-06-17 00:31:01 +02:00
  • fa43f69dfd docs(audit): DQ-3 content fidelity verified clean — assessment axis complete Tarik Moussa 2026-06-17 00:23:16 +02:00
  • ff03443af4 feat(ingest): recover missing metadata + abstracts from arXiv/S2/Crossref (DQ-2) Tarik Moussa 2026-06-17 00:12:59 +02:00
  • 51dc74470c feat(ingest): supplement citations from Semantic Scholar when OpenAlex is empty (DQ-1) Tarik Moussa 2026-06-16 23:51:48 +02:00
  • 553ae3995d fix(search): cast query embedding to ::vector in search paper (DQ-4) Tarik Moussa 2026-06-16 23:51:38 +02:00
  • d64edfc324 Merge pull request 'fix: audit remediation Wave 3 — M-1, MED/LOW sweep, ingest robustness + D-1 close-out' (#14) from fix/audit-wave-3 into main user2595 2026-06-16 04:53:59 +00:00
  • 07929bd4fe Merge pull request 'fix: audit remediation Wave 2 — identity normalization C-7, C-10 (+ T-3, ADR D-1)' (#13) from fix/audit-wave-2 into main user2595 2026-06-16 04:53:40 +00:00
  • dca0dd793f Merge pull request 'fix: audit remediation Wave 1 — correctness fixes C-1, C-4, C-11 (+ doc fixes)' (#12) from fix/audit-wave-1 into main user2595 2026-06-16 04:53:19 +00:00
  • fb06c6f65a Merge pull request 'docs+chore: audit findings + tooling (base of the remediation stack)' (#15) from audit/loop-1 into main user2595 2026-06-16 04:52:58 +00:00
  • 64e3ee188c docs(audit): data-quality findings + handoff for a fresh session Tarik Moussa 2026-06-15 22:45:05 +02:00
  • e1c0a98262 fix(ingest): make both upsert UNIQUE gaps visible (audit C-15 follow-up) Tarik Moussa 2026-06-15 22:25:30 +02:00
  • 1f677211f7 docs(adr): finalise D-1 — re-measure F-15 spike on canonical corpus Tarik Moussa 2026-06-15 22:13:56 +02:00
  • b52303d22b fix(ingest): retry on bibkey UNIQUE collision (audit C-15) Tarik Moussa 2026-06-15 22:11:14 +02:00
  • 45c3b16dd7 docs: clarify sparse-not-wired and cite-boost tie-breaker (audit C-14, U-2) Tarik Moussa 2026-06-15 21:47:40 +02:00
  • f422b0fb84 fix(wiki): token-boundary grounding match + sharper ref-list filter (C-6, R-2) Tarik Moussa 2026-06-15 21:46:02 +02:00
  • 92928bab99 feat(cli): graph-report titles + JSON warning; wire recommend/cocited (U-1,U-3,U-4,C-9) Tarik Moussa 2026-06-15 21:43:19 +02:00
  • c7fae5c877 fix(sources): drop no-op try/except + url-encode S2 ids (audit R-4, R-5) Tarik Moussa 2026-06-15 21:39:48 +02:00
  • b4fbd7930e feat(config): wrap credentials in SecretStr (audit S-4) Tarik Moussa 2026-06-15 21:32:13 +02:00
  • e7b854db10 fix(wiki): make grounding tokenizer robust to punctuation (audit R-1) Tarik Moussa 2026-06-15 16:37:30 +02:00
  • 77f9d79da0 fix(mcp): real cited bibkeys in wiki_read + constant-time token compare Tarik Moussa 2026-06-15 16:34:28 +02:00
  • c4106b0f51 fix(synthesis): make coverage-map gaps opt-in (audit R-8) Tarik Moussa 2026-06-15 16:32:01 +02:00
  • cfbfa53398 fix(wiki): protect math in cross-refs; drop no-op DB call; fix mark order Tarik Moussa 2026-06-15 16:24:22 +02:00
  • ae5e9a528b docs: document 'codex migrate' + MIGRATION_DATABASE_URL as the schema-sync path Tarik Moussa 2026-06-15 16:20:54 +02:00
  • 10ff5ba3ab chore(infra): point migration helper preflight at 'codex migrate' Tarik Moussa 2026-06-15 15:54:25 +02:00
  • cdb7d254df feat(db): idempotent schema + 'codex migrate' command (audit M-1, T-2) Tarik Moussa 2026-06-15 15:53:25 +02:00
  • b4cf924bf1 docs(audit): record M-1 migration incident + privilege dimension Tarik Moussa 2026-06-15 15:45:39 +02:00
  • 9f172f5c68 fix(infra): preflight schema check before TRUNCATE in migration helper Tarik Moussa 2026-06-15 15:37:35 +02:00
  • 1ff8052b69 chore(infra): add canonical-id re-ingest migration helper (audit C-7) Tarik Moussa 2026-06-15 12:17:17 +02:00
  • ce75bc02e5 docs(adr): audit addendum to ADR-F15 — ID mismatch resolved (audit D-1) Tarik Moussa 2026-06-15 12:10:46 +02:00
  • 8e2f9e037d fix(ingest): key formulas/figures on papers.id, not filename stem (audit C-10) Tarik Moussa 2026-06-15 12:09:38 +02:00
  • c36e7c6ee7 fix(ingest): canonicalise papers.id to the caller id (audit C-7, T-3) Tarik Moussa 2026-06-15 12:06:53 +02:00
  • 6193f5630d docs: fix two doc/code contradictions (audit D-2, D-4) Tarik Moussa 2026-06-15 11:13:52 +02:00
  • eee147275a fix(discover): resolve cited_id via shared resolver (audit C-1) Tarik Moussa 2026-06-15 11:12:55 +02:00
  • 3092f1814e fix(synthesis): namespace lead ids by kind to stop collisions (audit C-11) Tarik Moussa 2026-06-15 11:09:24 +02:00
  • 115bb63f2d fix(wiki): quarantine zero-citation pages instead of publishing (audit C-4) Tarik Moussa 2026-06-15 11:06:12 +02:00
  • 8af9f9ad0a docs(audit): runtime-validate C-4 & C-11, close nougat.py review Tarik Moussa 2026-06-15 10:57:24 +02:00
  • 33fb91d478 docs(audit): live-DB verification — confirm C-7 & C-1, refine M-1 Tarik Moussa 2026-06-15 10:40:46 +02:00
  • 2d1b0b5e23 docs(audit): complete whole-repo audit (parts D-G + exec summary) Tarik Moussa 2026-06-15 10:14:50 +02:00
  • 7682ff112f docs(audit): whole-repo audit parts A-C (grounding, sources, parsing) Tarik Moussa 2026-06-15 09:23:56 +02:00
  • a0ad069b1d docs(audit): record development-loop-1 findings Tarik Moussa 2026-06-15 08:43:54 +02:00
  • 0c89eebeb8 chore(ops): ignore .env.* profiles + add batch ingest script Tarik Moussa 2026-06-15 08:43:53 +02:00
  • 9726042964 style: ruff format normalization to line-length=100 Tarik Moussa 2026-06-15 08:43:53 +02:00
  • 5dfa6e4bae feat(graph): paper_identifiers table — multi-ID alias support Tarik Moussa 2026-06-15 07:42:21 +02:00
  • 664600fdc7 fix(graph): resolve cited_id to papers.id via openalex_id JOIN Tarik Moussa 2026-06-15 07:15:21 +02:00
  • a1a6f4590b docs(adr): ADR-F15 literature graph — GO, Bobenko+Springborn rank 7 Tarik Moussa 2026-06-15 07:11:37 +02:00
  • 52d49dcfc6 docs(adr): ADR-F15 literature graph — GO, Bobenko+Springborn rank 7 feat/F-15-literature-graph Tarik Moussa 2026-06-15 07:11:37 +02:00
  • 2c476dc038 Merge pull request 'feat(F-15): citation graph — PageRank, coupling, co-citation, CLI' (#11) from feat/F-15-literature-graph into main user2595 2026-06-15 03:18:22 +00:00
  • 1f1e0e82b3 fix(F-15): address Opus review findings (CRITICAL + WARN) Tarik Moussa 2026-06-15 03:40:30 +02:00
  • fd51b70000 feat(F-15): citation graph — PageRank, coupling, co-citation, CLI Tarik Moussa 2026-06-15 03:31:36 +02:00
  • 9647897173 feat(F-16): chunk quality gate + section classification user2595 2026-06-15 01:22:44 +00:00
  • b61d84d4bd feat(F-16): chunk quality gate + section classification feat/F-16-chunk-quality Tarik Moussa 2026-06-15 03:22:11 +02:00
  • 1a9afb4433 Merge pull request 'fix(ingest): D-04 bibkey heuristic + D-05a/b/c regression tests + __main__' (#9) from fix/d04-d05-ingest-fixes into main user2595 2026-06-15 01:07:35 +00:00
  • b87087d3c6 fix(ingest): D-04 bibkey heuristic + D-05a/b/c regression tests + __main__ Tarik Moussa 2026-06-15 03:07:06 +02:00
  • 335ca2c923 docs(adr): ADR-F13 synthesis engine — GO + zero-fact-leak enforcement (N-02) Tarik Moussa 2026-06-15 01:54:53 +02:00
  • 36a5d9566a Merge pull request 'fix(docs): correct provenance CLI command names in README (D-01)' (#8) from fix/cli-readme-drift into main user2595 2026-06-14 23:51:07 +00:00
  • b12789a47f fix(docs): correct provenance CLI command names in README (D-01) Tarik Moussa 2026-06-15 00:49:50 +02:00
  • 1bec18ae82 Merge pull request 'fix(ingest): support .txt sources + fix citing_id FK violation' (#7) from fix/ingest-txt-and-citing-id into main user2595 2026-06-14 13:56:04 +00:00
  • 5025ff3a88 fix(ingest): support .txt sources + fix citing_id FK violation Tarik Moussa 2026-06-14 15:54:05 +02:00
  • 6c28446644 Merge pull request 'fix(mcp): disable DNS rebinding protection for LAN HTTP transport' (#6) from fix/mcp-transport-security into main user2595 2026-06-14 12:55:59 +00:00
  • d29dcf2fb5 fix(mcp): disable DNS rebinding protection for LAN HTTP transport Tarik Moussa 2026-06-14 14:55:29 +02:00
  • 1d1203b225 Merge pull request 'fix(sources): resolve arXiv IDs via arXiv DOI (OpenAlex /works/arxiv: 404s)' (#5) from fix/openalex-arxiv-doi into main user2595 2026-06-14 12:31:33 +00:00
  • 50e6bb8210 fix(config): remove duplicate F-13 settings block (merge artefact) Tarik Moussa 2026-06-14 08:59:05 +02:00
  • b2ef8f2575 docs(mcp): client registration (Claude Code / Desktop) Tarik Moussa 2026-06-14 08:23:34 +02:00
  • b803811701 feat(mcp): FastMCP server exposing read-only KB tools (stdio) Tarik Moussa 2026-06-14 08:23:22 +02:00
  • 59763105a5 fix(synthesis): enforce non-empty provenance + fix gap lead + clean index docstring Tarik Moussa 2026-06-14 08:52:35 +02:00
  • 8cf0cc7e01 feat(synthesis): Lead/Provenance model + grounded leads + conjecture generator Tarik Moussa 2026-06-14 08:37:42 +02:00
  • 967e6baa59 feat(wiki): ADR-F12-wiki-compile.md — GO decision + grounding measurement Tarik Moussa 2026-06-14 03:54:42 +02:00
  • 7d338d29ca fix(wiki): sentence-level claim granularity + content-5-gram grounding guard Tarik Moussa 2026-06-14 03:53:39 +02:00
  • 408e4886bb fix(wiki): basic conflict detection in CompileReport Tarik Moussa 2026-06-14 03:51:30 +02:00
  • ef46d14aed fix(wiki): protect inline-code spans from cross-ref injection Tarik Moussa 2026-06-14 03:49:06 +02:00