Commit Graph

95 Commits

Author SHA1 Message Date
336962715a fix(rag): S3 NLI unavailable — hypothesis char budget + token-count fallback retry
S3 dogfood query 의 `nli_model_unavailable` consistent fail root cause = mDeBERTa-v3 tokenizer 의 `OnlyFirst` strategy + 949-token hypothesis. 기존 char-budget 단독 fix 의 KR-extreme density 미해결 → token-count fallback retry + RC1-residual trait dispatch 정합.

핵심 변경:
- kebab-nli::NliVerifier: `hypothesis_token_count(&str) -> Result<usize>` trait method 추가 (default `Ok(0)` backward-compat). `OnnxNliVerifier` 가 *trait impl block* 안에서 real mDeBERTa tokenize override — vtable 등록 보장 (round-3 critic RC1-residual closure).
- kebab-rag::pipeline: `MAX_NLI_HYPOTHESIS_CHARS_INITIAL = 1200` + `MAX_NLI_HYPOTHESIS_CHARS_MIN = 150` const + `pub(crate) fn truncate_chars` pure-fn + `pub fn truncate_hypothesis_for_nli_with_budget` retry helper (char budget 반감 retry, min floor 시 graceful unavailable). step 8.5 hook 의 callsite explicit `match` + `return self.refuse_nli_model_unavailable` 패턴 (`?` 금지 — round-2 plan critic CRITICAL #1 closure).
- SpyNliVerifier 신규 helper (closure score_fn + hypothesis_token_count_fn, 2-arg constructor).
- §5.1 의 2 ignored test (EN-long err + vtable dispatch RC1-residual pin) + §5.2 의 4 boundary test (truncate_chars) + §5.3 의 3 mock multi-hop test (long_en_grounded / long_kr_retries / unrelenting_fallback). +7 new tests (2 ignored default skip).
- tasks/HOTFIXES.md 신규 dated entry `## 2026-05-26 — S3 NLI unavailable ...` — Symptom / Root cause / Action / Amends 4-block.
- spec + plan (`docs/superpowers/{specs,plans}/2026-05-26-s3-nli-model-unavailable-diagnose-*.md`) — 4 round spec + 3 round plan OMC reviewer ACCEPT 산출물.

검증:
- cargo test -p kebab-nli -j 1 → 11/11 pass + 7 ignored default skip.
- cargo test -p kebab-rag -j 1 → 19+3+3+... 전체 pass + 3 new mock + 4 new boundary.
- cargo test --workspace --no-fail-fast -j 1 → **1313 pass (+7 new)**, 0 failed. 회귀 0 (HOTFIX #15 이미 fixed, no remaining flaky).
- cargo clippy --workspace --all-targets -j 1 -- -D warnings clean (type_complexity allow on Arc<dyn Fn> type aliases).

KR safe (token-count retry path) + graceful fallback (min floor 시 기존 unavailable wire 유지, regression 0). Wire 영향 없음 (additive trait method). Cargo bump 불필요.

Refs:
- spec: docs/superpowers/specs/2026-05-26-s3-nli-model-unavailable-diagnose-spec.md (4 round APPROVE — analyst → critic + verifier × 4 rounds)
- plan: docs/superpowers/plans/2026-05-26-s3-nli-model-unavailable-diagnose-plan.md (3 round ACCEPT — planner → critic-plan + verifier-plan × 3 rounds)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-26 09:12:21 +00:00
a210bf5d52 docs(rag): HOTFIX #15 spec + plan (3 round OMC reviewer approve)
OMC team `hotfix-15-mcp-flaky` 의 spec + plan 작성 + 리뷰 산출물.

- spec: analyst 가 진단 (root cause = PR-7 probe-first 가 PR-5 test 의 stale empty-KB contract 와 mismatch) + Option A 권장 (test-only fix). 3 round review (critic + verifier): CRITICAL C1 (fixture/query FTS5 0 hits) + MAJOR M1/M2 + 등 closure.
- plan: planner 가 7 steps + subagent dispatch task 작성. 3 round review (critic-plan + verifier-plan): empirical SQLite REPL 검증, level-1 dated entry placement, actual KebabHandler/KebabAppState pattern 정합.

implementation = 429287f commit (executor).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-26 06:52:04 +00:00
98cf4e8a04 chore(release): bump version 0.17.2 → 0.18.0 + cut fb-41 multi-hop
v0.18.0 cut PR. fb-41 multi-hop RAG + NLI verification 의 user-visible surface (PR #176-180) + post-PR9 cleanup/refactor (PR #181) ship 마무리.

## 변경 사항

### Version
- workspace `Cargo.toml`: 0.17.2 → 0.18.0. Cargo.lock 자동 cascade (24 kebab-* crate 모두 0.18.0).

### Frozen design contract
- `docs/superpowers/specs/2026-04-27-kebab-final-form-design.md`:
  - §3.8 RAG types — RefusalReason 에 NliVerificationFailed + NliModelUnavailable + MultiHopDecomposeFailed 추가 + Multi-hop RAG + NLI verification 의 ask_multi_hop facade + step 8.5 NLI hook + HopRecord / VerificationSummary 명시.
  - §9 versioning rules 표 — nli_model_version row 신규 (선택 — v0.19+ second adapter 시 wire surface candidate).

### Status transitions
- `docs/superpowers/specs/2026-05-25-p9-fb-41-finalize-spec.md`: status approved-by-team → completed.
- `docs/superpowers/plans/2026-05-25-p9-fb-41-finalize-plan.md`: status approved-by-team → completed (spec_status 도).

### User-facing docs
- `README.md`: 명령 표의 `kebab ask` row 에 `--multi-hop` flag + NLI 옵션 안내 한 단락 (mDeBERTa-v3 XNLI 280 MB 자동 다운로드 / RAM peak ~7-8 GB / threshold tuning 0.5 prod / 0.0 disable).
- `docs/SMOKE.md`: `[rag] nli_threshold = 0.0` config 예시 + 활성화 절차 + first-run download + RAM 권장 inline 안내.

### Handoff + dashboard
- `HANDOFF.md`: 한 줄 요약 의 현재 version 0.17.2 → 0.18.0. v0.18.0 cut entry 추가 (fb-41 multi-hop + NLI + cleanup ship). Component 카운트 단락에 fb-41 PR-9 의 kebab-nli + ask_multi_hop 추가 명시. 머지 후 결정 절 맨 위에 v0.18.0 fb-41 entry 신규.
- `tasks/INDEX.md`: p9-fb-41  머지 (v0.18.0). v0.18.0 subsection 신규 — PR #176-181 의 6 sub-PR + cleanup 각 한 줄 요약.

## 비범위 / 별 작업
- HOTFIXES.md 의 fb-41 entry 는 이미 PR #180 (PR-9d closure) 에서 작성 완료 — 본 cut PR 에서 추가 anchor 불필요.
- SKILL.md 의 v0.18+ NLI 안내는 이미 PR-9c-2 에서 inline 추가 완료.

## 검증
- `cargo check --workspace -j 1` 통과 (모든 24 crate v0.18.0 확인).
- frozen design 의 RefusalReason enum 확장이 kebab-core 의 production code 와 정합 (PR-9c-1 시점부터 동일 variants 있음).

Wire 영향: 없음 (additive minor 는 PR-9c-1 에서 이미 ship, 본 commit 은 documentation cascade only).
Behavior 영향: 없음.

머지 후 `gitea-release v0.18.0` 으로 tag + release notes 작성.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-26 05:18:08 +00:00
44fbffff26 docs(rag): fb-41 PR-9 spec + plan — NLI verification + v0.18.0 cut
fb-41 multi-hop RAG 의 dogfood S7 hallucination root cause = LLM-self-judge ceiling.
대응 = NLI-based post-synthesis verification (mDeBERTa-v3 XNLI, 280 MB ONNX).

산출물:
- docs/superpowers/specs/2026-05-25-p9-fb-41-finalize-spec.md (review_round=5,
  4 OMC reviewer APPROVE: 1 CRITICAL + 9 MAJOR + 3 MINOR → 1 NIT carry-forward).
- docs/superpowers/plans/2026-05-25-p9-fb-41-finalize-plan.md (plan_review_round=3,
  4 OMC reviewer APPROVE: 15 issues → 0 actionable).

5 sub-PR (PR-9a~9d) + cut PR. 작업 21-31h / wall time 28-44h.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-25 21:22:20 +00:00
7150c376bb feat(rag): fb-41 PR-3a — HopRecord wire + RagCfg multi-hop knobs
PR-3 의 분할 첫 PR. wire additive (HopRecord + HopKind + Answer.hops
field) + RagCfg 의 multi_hop_* 3 노브. RAG pipeline 동작 미변경 —
모든 Answer literal 의 `hops = None`. PR-3b (후속) 가 ask_multi_hop
의 happy path 에서 dynamic decide loop 구현 + hops trace 채움.

분할 이유: 원래 PR-3 가 wire + cfg + decide loop + ScriptedLm +
helper refactor + 5+ tests 단일 PR 였는데 ~1500 줄 단일 patch 가
review 부담 + 회기 위험 ↑. additive foundation 부터 ship 후 decide
loop 별 PR — 사용자 결정 (2026-05-25).

- `kebab_core::HopRecord` (iter, kind, sub_queries,
  context_chunks_added, forced_stop, llm_call_ms) + `HopKind`
  (Decompose / Decide / Synthesize) — wire-additive shape.
- `kebab_core::Answer.hops: Option<Vec<HopRecord>>` —
  `#[serde(default, skip_serializing_if = "Option::is_none")]`,
  single-pass / refusal path 는 None, PR-3b 의 multi-hop happy
  path 가 Some.
- `kebab_config::RagCfg` 에 3 신규 노브:
  - `multi_hop_max_depth: u32` (default 3)
  - `multi_hop_max_sub_queries_per_iter: u32` (default 5)
  - `multi_hop_max_pool_chunks: u32` (default 30)
  3 모두 `#[serde(default)]` + env override
  (`KEBAB_RAG_MULTI_HOP_MAX_*`) + legacy parse 핀
  (`LEGACY_PRE_TIMEOUT_TOML` 공유).
- 9 Answer literal site (pipeline.rs ×6 + kebab-cli + kebab-tui
  tests + kebab-eval test) 에 `hops: None` 명시 추가. exhaustive
  field check 가 자동 guard — 빠진 site 시 compile fail.
- plan 의 PR-3 단락 → PR-3a / PR-3b 분할 명시 + scope 정정.

Tests (163 passing across kebab-config + kebab-core + kebab-rag):
- 5 신규 multi-hop knob test (default / env override / legacy parse).
- 기존 50+57+31+19+3+3 test 모두 hops:None 추가 후도 통과.

Wire 영향: `answer.v1` 의 optional `hops` 필드 — `skip_serializing_
if = None` 이라 single-pass response 에 emit 안 됨. wire breaking
아님, JSON Schema 갱신은 PR-3b 또는 PR-4 (실제 emit 시점).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-25 07:15:01 +00:00
192da45dbf chore(rag): PR #167 회차 1 리뷰 반영
- `parse_decompose_response_drops_partial_empty_keeps_valid` 신규
  회귀 핀 — `["", "valid q", "  "]` → `["valid q"]` (trim+filter
  chain 동작 pin).
- `multi_hop_decompose` 의 `stop: Vec::new()` 옆 doc comment 추가 —
  의도 명시 (instruction-following 모델 기대 + prose 추가 시
  MultiHopDecomposeFailed refusal 가 policy). 회차 1 question 의
  답변.
- plan 의 PR-3 implementation order 에 회차 1 carry-over 추가:
  1) ask + ask_multi_hop 의 §4-§9 mirror → 공통 helper 추출,
  2) decompose template 의 substitution corner case → format!
     named arg 으로 교체.

회차 1 의 다른 suggestion (mirror refactor, substitution corner
case, history block helper) 는 PR-3 합리적 timing 으로 plan 에
명시 — 회차 2 reply 에 정리.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-25 06:49:21 +00:00
624b44c46b chore(eval): PR #166 회차 1 리뷰 반영
- `mh-s-004` 의 `must_contain: ["i"]` 한 글자 → `["INSERT", "i 입력모드"]`
  보강. trigram 0-hit + noise 매칭 위험 해소.
- 3 question 영어 변경 (`mh-c-005` / `mh-i-001` / `mh-s-002`) — fixture
  의 lang 다양성 mix (12 ko + 3 en). 영어 dogfood 시 measurement gap
  회피.
- plan 의 PR-1 단락이 outdated (kebab-eval crate 미survey 단계 작성 →
  실제 PR 와 deviation). actual 변경 명시 + 초안 대비 deviation 명시.

회차 1 의 다른 2 suggestion (mh-c-002 의 `v0.17.2` hard-coded, 15
question / 5-per-bucket 회귀 핀의 frozen size) 은 baseline anchor 의도
적 freeze — 회차 2 reply 에 명시.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-25 06:26:15 +00:00
caf690dc72 feat(eval): fb-41 multi-hop golden set + spec/plan
PR-1 of fb-41 multi-hop RAG (spec: docs/superpowers/specs/2026-05-25-
p9-fb-41-multi-hop-rag-design.md, plan: docs/superpowers/plans/2026-
05-25-p9-fb-41-multi-hop-rag.md).

XL 작업의 첫 PR — baseline 측정 anchor 만 추가. RAG pipeline 미변경,
fixture file + parse 회귀 핀.

사용자 결정 4 axis (2026-05-25):
- approach: query decomposition (LLM 서브-질문)
- trigger: explicit `--multi-hop` flag
- MVP scope: dynamic N-hop (LLM 이 depth 결정, decompose seed +
  ReAct-style decide loop hybrid)
- eval: multi-hop golden set 먼저 (본 PR)

본 PR:
- `fixtures/multi_hop_golden.yaml` 신규. 15 question (5 cross-doc +
  5 intra-doc + 5 single-fact negative). 기존 `GoldenQuery` struct
  그대로 사용 — 별 loader / type 변경 없음. `expected_chunk_ids`
  비어 있어 curator 가 `kebab ingest` 후 채울 수 있는 template
  형태. `must_contain` 으로 baseline 측정 가능 (P5-2 metric).
- `crates/kebab-eval/tests/loader.rs::loads_multi_hop_golden_fixture`
  신규 회귀 핀. fixture parse OK + 15 question + 5/5/5 bucket
  분포 + 모든 question 에 must_contain 최소 1 개.

baseline 측정 protocol (별 run, commit 에 artifact 안 포함):
1. v0.17.2 binary 로 single-pass `kebab eval run --fixture
   multi_hop_golden.yaml` 실행
2. P@5, P@10, must_contain pass rate, citation_coverage 캡처
3. PR-3 (dynamic iter 머지) 후 동일 fixture + `multi_hop=true` 로
   재실행 → Δ 비교

PR 분할 6 단계 (plan 참조): PR-1 (본 PR — fixture only), PR-2
(RagPipeline::ask_multi_hop fixed depth=2), PR-3 (dynamic iter),
PR-4 (CLI flag + wire), PR-5 (MCP + SKILL.md), PR-6 (TUI toggle +
trace render). 마지막 PR 후 v0.18.0 cut.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-25 06:22:08 +00:00
fe123c0c6d test(A4): korean + english trigram matching at FTS level
3개 신규 unit tests in tests/fts.rs §7:

1. fts_trigram_korean_3char_substring_hits — Codex sqlite 3.45.1 검증
   동작 5개 assert pin: raw 3자 substring hit (충돌은/발생한),
   quoted phrase hit (\"해시 충돌\"/\"시 충\"), raw 해시충 0-hit (원문
   미존재).
2. fts_trigram_korean_short_query_zero_hit_pinned — 2자 한국어 query
   (충돌·키) 0-hit 회귀 감지. trigram 구조 변경 시 먼저 fail.
3. fts_trigram_english_substring_hits — substring recall 동작 변경
   pin (token→tokenizer, to 0-hit).

검증: cargo test -p kebab-store-sqlite --test fts → 13/13 PASS
(신규 3 + 기존 10).

Step 1c (multi-token 한국어 query e.g. \"해시 충돌\") 와 Step 5
(lexical BM25 snapshot 갱신) 는 Task A5 의 build_match_string()
재설계 후 진행.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 00:57:37 +00:00
753b1ff5e5 task(A4-step0): synthetic korean fixture for trigram tests
도그푸딩 실 한국어 위키 문서 (hash-table.md, 4512줄 mediawiki HTML,
CC-BY-SA) 는 크기·라이선스 부담으로 직접 commit 회피. 대신 도그푸딩
query 들 (해시 충돌·충돌은·시 충·해시충·충돌) 을 모두 cover 하는 합성
fixture 작성. trigram tokenizer 의 정확한 매칭 동작 (3자 substring
hit, 2자 0-hit, raw vs quoted phrase) 검증용.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 00:54:30 +00:00
8dcedc4b11 feat(p10-r2): V007 trigram migration + design §5.5 + fts diff-check
Task A2 + A3 한 묶음.

migrations/V007__fts_trigram.sql 신규:
- chunks_fts shadow 를 DROP + 재생성 (tokenize = trigram).
- chunks_ai/ad/au trigger 재생성 (V002 와 동일).
- chunks 에서 backfill INSERT — 사용자 re-ingest 불필요, V007 자동.
- V002 는 historical cold-upgrade replay 위해 그대로 유지.

design §5.5 갱신:
- verbatim block 의 tokenize 만 trigram 으로 교체.
- §5.5 본문 상단에 한국어 채택 사유 + trade-off (영어 lexical 변경,
  BM25 분포, 디스크 ~2-10x, contentless 아님) prose 한 단락 추가.

crates/kebab-store-sqlite/tests/fts.rs:
- fts_v002_matches_design_section_5_5_verbatim →
  fts_v007_matches_design_section_5_5_verbatim 으로 rename.
- extract_migration_5_5_verbatim_block() 의 include_str! path 를
  V007__fts_trigram.sql 로 변경. 주석/assertion msg V007 로.
- V002 cold-upgrade test 들 (fts_v002_backfill_*) 은 그대로 유지.

검증: cargo test -p kebab-store-sqlite --test fts → 10/10 PASS
(`fts_v007_matches_design_section_5_5_verbatim` 포함).

Codex round 1/2 의 design §5.5 contentless 정정·trigram tokenizer
채택 사유 명시 발견 반영.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 00:52:40 +00:00
8781c6112b task(A1): builder baseline + sqlite version + snapshot locations
Task A1 step 1-3 완료. plan A5 의 baseline 노트 슬롯 채움.

핵심 발견:
- build_match_string() (lexical.rs:177-200): trim → strip_single_quotes
  raw FTS verbatim / 그 외 whitespace split + escape_fts5_token (\"...\"
  + inner doubling) + space join (implicit AND).
- raw mode = single quote '...' 가 trimmed 전체 감쌈 (lexical.rs:167).
- SQLite: rusqlite 0.32 + libsqlite3-sys 0.30.1 bundled (in-tree, SQLite
  ~3.46.x) → trigram 사용 가능.
- Snapshot: tests/lexical.rs::lexical_snapshot_run_1 + tests/hybrid.rs::
  hybrid_snapshot_run_1 (KEBAB_UPDATE_SNAPSHOTS=1 로 regenerate).
  inline normalize_bm25_top_score 는 numerical 무관.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 00:47:24 +00:00
584247f1ea spec+plan(v0.17.0): korean trigram tokenizer + dogfood fixes
P10 도그푸딩 round 2 (2026-05-22) follow-up. SQLite FTS5 tokenizer
unicode61 → trigram 으로 교체해 한국어 lexical 검색 지원 + 작은
버그픽스 2 (C typedef-wrapped struct 미노출, code_lang_breakdown
집계 단위).

Codex + Gemini round 1/2/3 리뷰 반영:
- [r1] 2자 한국어 query 0-hit, build_match_string() multi-token 깨짐,
  contentless → shadow, parser_version cascade, BM25/heading_path/디스크
- [r2] same-workspace_path orphan purge (parser bump cascade 실제 동작),
  trigram 테스트 예시 sqlite 3.45.1 검증, builder 권장안 (whole phrase OR)
- [r3] SMOKE 시나리오 정정, TUI stale hint 방지, search_response.v1 hint
  필드, new purge helpers, single quote raw mode 통일, fixture 도입

PR 구성: PR-A (trigram + builder + 안내), PR-B (C typedef + orphan
purge), PR-C (stats + wire). 셋 머지 후 v0.17.0 release cut.

design: docs/superpowers/specs/2026-05-22-korean-trigram-tokenizer-design.md
plan:   docs/superpowers/plans/2026-05-22-korean-trigram-tokenizer.md

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 00:43:31 +00:00
438870ee25 docs(p10-1d): activate C + C++ in frozen design §10
P10 Tier 1 chunker family complete (Rust + Python + TS + JS + Go + Java +
Kotlin + C + C++).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 14:32:26 +00:00
a58d400abd docs(p10-1d): implementation plan (11 tasks A-K, subagent-driven)
Tasks: workspace deps / C extractor / C++ extractor / C chunker + snapshot /
C++ chunker + snapshot / ingest dispatch + tier3_fallback_cv extension /
2 smoke tests / frozen design §10 / docs sync / workspace test gate /
version bump 0.15.0 → 0.16.0 + gitea PR.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 13:15:22 +00:00
6bc7a83d3c docs(p10-3): activate Tier 3 in frozen design §10.1
Add p10-3 activation log entry for Tier 3 paragraph fallback chunker
(code-text-paragraph-v1) with shell direct routing and fallback wrapper
for invalid YAML / AST failures.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 11:39:49 +00:00
a8aa03042f docs(p10-3): implementation plan (9 tasks A-I, subagent-driven)
Tasks: tier2_shared visibility upgrade / Tier 3 chunker + 4 unit tests /
shell direct routing / Tier 1/2 fallback wrapper / 2 smoke tests / frozen
design §10.1+§10 / docs sync (6 files) / workspace test gate / version
bump 0.14.0→0.15.0 + gitea PR.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 11:16:55 +00:00
522ae7b8bc docs(p10-2): activate Tier 2 in code-ingest design §10.1 + §3.5 mappings
§3.5: add code_lang_for_path mappings xml / groovy / go-mod.
§10.1: add deactivation log entry for p10-2 (3 Tier 2 chunkers active).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 13:24:16 +00:00
5ce7f60932 docs(p10-2): implementation plan (11 tasks A-K, subagent-driven)
Branch feat/p10-2-tier2-resource. Tasks: serde_yaml dep / lang.rs basenames /
media.rs source-of-truth consolidation / 3 chunkers (k8s + dockerfile +
manifest) + tier2_shared helper / ingest dispatch / smoke tests / frozen
design §3.5+§10.1 / docs sync / version bump 0.13.0→0.14.0.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 12:55:36 +00:00
2d7a566624 docs(p10-1c-jk): README/HANDOFF/ARCHITECTURE/SMOKE/INDEX + design §10.1; chore: bump version 0.12.0 → 0.13.0
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 11:38:40 +00:00
1b19e33a4f docs(p10-1c-jk): task spec + implementation plan
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 10:27:13 +00:00
f95cd55484 docs(p10-1c-go): README/HANDOFF/ARCHITECTURE/SMOKE/INDEX + design §10.1; chore: bump version 0.11.1 → 0.12.0
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 10:02:21 +00:00
8b89961ada docs(p10-1c-go): task spec + implementation plan
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 08:58:45 +00:00
44813df052 docs(p10-1b): README/HANDOFF/ARCHITECTURE/SMOKE/INDEX + HOTFIXES; chore: bump version 0.7.0 → 0.8.0
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 01:48:06 +00:00
39b766ea59 docs(p10-1b): task spec + implementation plan
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 00:26:58 +00:00
80c2d31fb3 docs(p10-1a-2): README/HANDOFF/ARCHITECTURE/SMOKE/INDEX + HOTFIXES; chore: bump version 0.6.0 → 0.7.0
- README: note Rust .rs ingest active (code-rust-ast-v1), update Mermaid parse node + chunker labels, update supported formats note in Quick start and ingest command table; add code citation fields (symbol, code_lang, repo) and filter flags note
- HANDOFF: flip P10 row to note 1A-1  + 1A-2 PR open; add one-liner cross-link to HOTFIXES 2026-05-19 entries
- ARCHITECTURE: add kebab-parse-code node + edge (app → pcode, pcode → ptypes) to Mermaid graph; add directory tree entry; add code parser locked-in decision row (tree-sitter lives parser-side, design §6.3)
- SMOKE: add P10-1A-2 Rust code ingest section (ingest.code config keys, verification steps, known behaviors); add checklist item
- tasks/INDEX.md: flip p10-1A-1 to , update p10-1A-2 to 🟡 PR open
- tasks/p10/INDEX.md: same flips
- tasks/HOTFIXES.md: add two 2026-05-19 dated entries (AST_CHUNK_MAX_LINES constant vs config deviation + SourceType::Code deferred)
- tasks/p10/p10-1a-2-rust-ast-chunker.md: append two HOTFIXES cross-link lines in Risks/notes
- docs/superpowers/specs/2026-04-27-kebab-final-form-design.md §10.1: note p10-1A-2 surface activation
- Cargo.toml: version 0.6.0 → 0.7.0 (dogfooding-ready = minor bump trigger per CLAUDE.md)
- Cargo.lock: regenerated

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-19 22:48:11 +00:00
7a6a24ad10 feat(p10-1a-2): add MediaType::Code(lang) variant
TDD: red → green cycle confirmed. New `Code(String)` variant serializes
as `{"code":"rust"}` via serde `rename_all = "lowercase"`. All exhaustive
`match` sites updated (`media_label`, `ingest_one_asset` catch-all →
explicit or-pattern). Design §3.5 enum listing synced. Also fix
`/target` symlink gitignore pattern so integration-test binary lookup
via workspace-relative path works with CARGO_TARGET_DIR redirect.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-19 17:14:45 +00:00
9f3edb7e24 feat(p10-1a-2): add internal SourceSpan::Code variant + design §3.4 sync
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-19 15:52:01 +00:00
a08ed32199 docs(p10-1a-2): task spec + implementation plan
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-19 15:36:08 +00:00
th-kim0823
7961f8813d fix(p10-1a-1): PR review round 1 — doc inconsistencies
회차 1 review 의 4 건 actionable 모두 반영:

1. frozen design §2.1 의 code variant 예시에서 존재하지 않는 `repo` 필드 제거 + nested form 에서 actual wire (flat) 형태로 정리. 5 variant 의 nested-form illustrative example 은 그대로 두고, code variant 만 별도 block 으로 분리해서 actual wire 와 1:1 매칭. 또 위쪽 6 variant nested-form group 에서도 'code' 행 삭제 (정확한 contract 는 별도 block 에 있음).
2. §2.2 SearchHit 예시의 `repo: null, code_lang: null` + 'omitted when null' 주석 모순 제거 — 키 자체를 빼고 inline 주석으로 'markdown hit 에는 absent, 코드 hit 에서만 surface' 설명.
3. HANDOFF Phase row 식별자 `**10**` → `**P10**` (다른 row 와 일관성).
4. README synopsis 의 중복 `[--media code]` 제거 (`--media` 는 이미 위쪽에 한 번 있음, code 는 값 중 하나라 prose 에서 설명).

코드 변경 없음 — 모두 markdown 문서.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-15 18:24:15 +09:00
th-kim0823
7bbd2c0cbf docs(p10-1a-1): wire schema + frozen design + README/HANDOFF/SMOKE + task index
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-15 17:41:26 +09:00
th-kim0823
005a9011ea plan(p10-1a-1): code ingest framework implementation plan + spec wire-shape fix
21 task plan: kebab-core 도메인 타입 (Citation::Code variant, SearchHit repo/code_lang, IngestReport skip counters, Metadata extension), 새 kebab-parse-code crate (lang/repo/skip 모듈, gix dep), kebab-source-fs gitignore+blacklist 통합, kebab-config [ingest.code] 절, kebab-cli --repo/--code-lang flag, wire schema JSON 갱신, frozen design doc 갱신, README/HANDOFF/SMOKE 갱신, task index. 각 task 가 5-step TDD cycle (test fail → impl → pass → commit). 코드 chunker 는 1A-1 에 없음 — 1A-2 에서 추가.

spec 의 Citation::Code 예시가 기존 5 variants 의 flat wire 형태와 안 맞아서 (`code: {...}` 중첩이 아니라 top-level field) 같이 fix.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-15 14:31:22 +09:00
th-kim0823
c6d61b0b37 spec(p10): split Phase 1A into 1A-1 (framework) and 1A-2 (Rust chunker)
1A 가 들고 들어가는 *프레임워크 surface* (Citation `code` variant, SearchHit repo/code_lang, --media code / --code-lang / --repo filter, skip 정책, IngestReport 세분화, config 절, kebab-parse-code crate skeleton) 가 *언어 chunker 자체* 와 독립 검증 가능 — 1A-1 머지 후 기존 markdown corpus 의 wire 출력이 byte-level identical 한지 regression test 로 검증한 다음 1A-2 에서 Rust AST chunker 자체에 집중. binary version bump 트리거도 1A-2 로 미룸 (1A-1 은 wire additive minor + 사용자 surface 변경 없음).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-15 14:20:10 +09:00
th-kim0823
49487dc46b spec(p10): code ingest design — Tier 1 AST + Tier 2 resource + Tier 3 fallback
수십 개 git repo (한 부모 dir 아래) 를 corpus 로 확장. Tier 1 (Rust/Python/TS-JS/Go/Java/Kotlin/C/C++) 은 tree-sitter AST per-language chunker, Tier 2 (k8s manifest / Dockerfile / Cargo.toml 류) 는 resource-aware chunker, Tier 3 (shell / fallback) 는 paragraph + line-window. embedding 은 multilingual-e5-large 유지 — cross-corpus 검색 위해. Phase 1A (Rust) 부터 1D (C/C++) + Phase 2 (Tier 2) + Phase 3 (Tier 3) 순으로 진행. ignore 통합 (.gitignore honor + .kebabignore 추가 + 최소 built-in safety net), generated header sniff, size cap 으로 첫 도그푸딩 비용 차단. 새 Citation variant `code`, SearchHit 의 repo/code_lang 필드, --media code / --code-lang / --repo filter — 모두 additive minor.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-15 14:15:59 +09:00
th-kim0823
c62a8ff503 docs(fb-39b): design + HOTFIXES + new task spec + INDEX + README + SMOKE
Tasks 4 + 5: comprehensive doc update for embedding upgrade (multilingual-e5-large).

- design §5 + §9: update embedding_model / dimensions references (384 -> 1024)
- HOTFIXES: add fb-39b entry with user re-ingest procedure + backwards-compat notes
- tasks/p9-fb-39b-embedding-upgrade.md: new task spec (completed status)
- INDEX.md: add fb-39b row under RAG quality phase
- fb-39 task banner: append fb-39b link as lever implementation
- README: update config defaults + fastembed model size + embedding field docs
- SMOKE.md: append embedding upgrade verification section with e5-small -> e5-large sequence

Wire schema: no change (additive at config level, new table created by existing code).
Binary version: 0.6.0 -> 0.7.0 (cascade rule: embedding_model change = minor bump).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 23:28:48 +09:00
th-kim0823
d5321701ea plan(fb-39b): embedding upgrade implementation plan
5 tasks: kebab-embed-local resolve_model arm + check_dim test,
kebab-config defaults + TOML template flip, cross-crate fixture
sweep (likely no-op since most tests use provider=none), docs
(design + HOTFIXES + new task spec + INDEX), README + SMOKE
walkthrough.

Post-merge: 0.6 → 0.7 binary bump per CLAUDE.md cascade rule.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 23:02:37 +09:00
th-kim0823
2c3461c465 spec(fb-39b): embedding model upgrade design
- multilingual-e5-small (384 dim) → multilingual-e5-large (1024 dim)
- Cascade: embedding_version bump → fb-23 incremental ingest
  re-embeds all chunks
- Migration policy: dim mismatch detection at LanceVectorStore::open
  → error.v1 (code = embedding_dim_mismatch) + hint
  "kebab reset --vector-only && kebab ingest"
- Config defaults flip (model + dimensions). User TOML pinning small
  preserves backwards-compat
- bge-m3 deferred (fastembed enum 미포함, UserDefinedEmbeddingModel
  ONNX path 별도)
- Release trigger: 0.6 → 0.7 minor bump per CLAUDE.md cascade rule

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 22:59:03 +09:00
th-kim0823
f00fb376fe docs(fb-39): golden header + design §10.3 eval + spec status + INDEX
Strengthen fixtures/golden_queries.yaml header with precision_at_k_chunk
explanation + measurement guidance. Add §10.3 Eval metrics section to
frozen design documenting retrieval metrics (hit@k, MRR, recall@k_doc,
P@k_chunk) + groundedness metrics. Flip p9-fb-39 spec status from open
→ completed (eval foundation only, lever deferral noted). Update
tasks/INDEX.md fb-39 row mirror to fb-42 (merged, deferred note).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 22:35:15 +09:00
th-kim0823
f303c76f52 plan(fb-39): eval foundation implementation plan
4 tasks: AggregateMetrics.precision_at_k_chunk field + serde
backwards-compat, compute aggregation in loop with 5 unit tests,
golden YAML header doc strengthening, design §11 + INDEX + status
flip.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 22:19:44 +09:00
th-kim0823
cd5b1e3bfc spec(fb-39): eval foundation design (P@k metric)
- AggregateMetrics 에 precision_at_k_chunk: BTreeMap<u32, f32>
  (P@5, P@10) 추가, binary relevance via expected_chunk_ids
- Denominator = k 고정 (hits.len() < k 도 precision 손실 간주)
- Empty expected_chunk_ids query 는 skip (hit_at_k 동일 정책)
- Lever 적용 (chunk policy / RRF / cross-encoder / embedding) 은
  본 spec 범위 외 — fb-39b 이후 별도 task
- Golden set schema 무변경, shipped fixtures 헤더 주석만 강화

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 22:05:09 +09:00
th-kim0823
441f1192ee docs(fb-42): wire schema + README + SMOKE + design + SKILL + INDEX
- Add bulk_search_item.v1 + bulk_search_response.v1 wire schemas
- Register both in WIRE_SCHEMAS const
- README: --bulk flag mention + MCP tool list 7→8 (bulk_search)
- SMOKE: bulk multi-query walkthrough (CLI + MCP equivalent)
- Design §2.2: Bulk multi-query (fb-42) subsection (additive minor)
- SKILL: mcp__kebab__bulk_search section + tool table row
- Task spec status open→completed, banner replaced
- INDEX: fb-42 row 머지 (rerank hint deferred)
- Fix: missed Capabilities {bulk_search} in cli wire.rs test (Task 7 leftover)
- Fix: missed tools.len() 7→8 in cli_mcp_smoke (Task 5 leftover)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 21:07:36 +09:00
th-kim0823
de9016fe16 plan(fb-42): bulk multi-query implementation plan
8 tasks: kebab-core types, kebab-app bulk_search_with_config facade
(cap 100 + per-query error policy), CLI --bulk flag + stdin ndjson +
output stream, CLI integration tests, MCP bulk_search tool +
registration + tools_list count bump, MCP integration tests,
capability flag, wire schemas + README + SMOKE + design + SKILL +
status flip.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 20:10:39 +09:00
th-kim0823
35df15df99 spec(fb-42): bulk multi-query design (rerank hint deferred)
- CLI: kebab search --bulk + stdin ndjson → stdout per-query ndjson
- MCP: 신규 kebab__bulk_search tool + JSON envelope (results + summary)
- Sequential for-loop, App instance 재사용 (cache amortize)
- Per-query error policy: continue + per-item error.v1
- Limits: queries.len() <= 100
- Capability flag bulk_search 신규
- Rerank hint 별도 task (fb-39 cross-encoder 설계 후)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 20:05:27 +09:00
th-kim0823
600c6182fc docs(fb-40): rag-v2 prompt + README + design + SKILL + INDEX
- README: [rag] prompt_template_version default rag-v2 + V2 강화 3 규칙
- design §7: rag-v2 본문 + V1 legacy note
- SKILL.md: mcp__kebab__ask 응답 행태 변화 안내
- task spec: status open → completed, design + plan 링크
- INDEX: fb-40  머지 (2026-05-10)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 19:37:28 +09:00
th-kim0823
6d6eb442be plan(fb-40): fact-grounded answer implementation plan
6 tasks: SYSTEM_PROMPT_RAG_V2 + system_prompt_for helper, pipeline
dispatch wiring, config default flip rag-v1 → rag-v2, test fixture
cleanup, integration tests (rag-v1 / rag-v2 / unknown via
CapturingLm wrapper around MockLanguageModel), docs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 18:58:35 +09:00
th-kim0823
28d3250546 spec(fb-40): fact-grounded answer design
- rag-v1 → rag-v2 system prompt with 3 신규 규칙 (verbatim span 인용 자도 /
  학습 지식 동원 금지 / 추측 금지)
- system_prompt_for(version) helper dispatch in pipeline
- config default prompt_template_version "rag-v1" → "rag-v2", V1 legacy
  kept for backwards-compat
- Lever C (pre-LLM gate) already shipped (RefusalReason::ScoreGate),
  out of scope here

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 18:55:05 +09:00
th-kim0823
c864bd007f docs(fb-38): wire schema + README + design + SKILL + INDEX 2026-05-10 18:21:55 +09:00
th-kim0823
56f20b7235 plan(fb-38): score semantics implementation plan
7 tasks: kebab-core ScoreKind enum + SearchHit field, lexical Bm25
labeling, vector Cosine, hybrid Rrf + search_with_trace pass-through,
cross-crate SearchHit literal cleanup, CLI integration test, docs
(wire schema + README + design + SKILL + INDEX).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 17:45:57 +09:00
th-kim0823
0359bd9682 spec(fb-38): score semantics design
- search_hit.v1 에 optional score_kind 필드 (rrf | bm25 | cosine)
- LexicalRetriever → Bm25, VectorRetriever → Cosine, HybridRetriever → Rrf
- fb-37 search_with_trace 의 mode-dispatch hits 는 underlying retriever 의
  score_kind 그대로 보존
- README + design §4 + SKILL 에 RRF 수식 전체 + "ranking signal, NOT confidence"
  안내, agent 용 trust threshold 는 nested retrieval.{lexical,vector}_score
- additive minor wire — schema bump 없음

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 17:40:47 +09:00
th-kim0823
fb31befef1 plan(fb-37): trace + stats implementation plan
10 tasks: kebab-core types, store breakdowns/index_bytes helpers,
extended CountSummary + Stats wire mirror, HybridRetriever
search_with_trace, App SearchResponse.trace threading, CLI --trace
flag, integration tests, MCP SearchInput.trace, TUI TracePopup,
docs (wire schema + README + SMOKE + INDEX + SKILL).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-10 12:14:26 +09:00