Every commit scored for change-risk against this repo's own history, so 'elevated' means elevated here rather than on some global curve.
Needs review
48 commits sit in this repo's top risk tercile, which is 33% of the 146scored. The cut is drawn against this codebase's own history rather than a global curve, so a quiet repo still fills its top band, and here it starts at 8.8 out of 10. What pushes a commit up is size and spread together: a large change confined to one area scores below a smaller one scattered across a dozen files.
Commit categories over time, read off the subject line. Fixes carry the accent because that is the series this chart exists to show.
How this repo's commit mix shifts over time.
Ranked by change-risk, highest first. Priority is a tercile of this repo's own distribution, so a quiet repo still fills its top band.
| # | Commit | Author | When | Lines | Review priority | Top driver |
|---|---|---|---|---|---|---|
| 1 | 3a0058d6Revert "chore: reset to clean LLM-setup-only state" | cmblir | 4mo ago | +6.4K -1 | 98thElevated | more lines added than baseline |
| 2 | 3be7c7bfindex: rebuild (hierarchical) | cmblir | 4mo ago | +988 -24 | 98thElevated | more lines added than baseline |
| 3 | b585f3f0index: rebuild (hierarchical) | cmblir | 4mo ago | +988 -24 | 98thElevated | more lines added than baseline |
| 4 | b3139b9dindex: rebuild (hierarchical) | cmblir | 4mo ago | +988 -24 | 98thElevated | more lines added than baseline |
| 5 | fab9e43binit: LLM Wiki 프로젝트 초기화 | cmblirclaudeassisted | 4mo ago | +1.9K -0 | 98thElevated | more lines added than baseline |
| 6 | 54c5f923feat: README 전면 재작성 + 가이드 스트리밍 + 떠다니는 Claude 캐릭터 챗봇 | cmblirclaudeassisted | 4mo ago | +1.0K -371 | 94thElevated | more lines added than baseline |
| 7 | 75791a87feat: full README rewrite + streaming guide + floating character chatbot | cmblir | 4mo ago | +1.0K -371 | 94thElevated | more lines added than baseline |
| 8 | bdaa32a6feat: full README rewrite + streaming guide + floating character chatbot | cmblir | 4mo ago | +1.0K -371 | 94thElevated | more lines added than baseline |
| 9 | b5680209feat: i18n (EN/KO 토글) + Karpathy-flavored 기능 7종 | cmblirclaudeassisted | 4mo ago | +859 -148 | 94thElevated | more lines added than baseline |
| 10 | b0f4af42feat: i18n (EN/KO toggle) + 7 Karpathy-flavored features | cmblir | 4mo ago | +859 -148 | 94thElevated | more lines added than baseline |
| 11 | 9c953323feat: i18n (EN/KO toggle) + 7 Karpathy-flavored features | cmblir | 4mo ago | +859 -148 | 94thElevated | more lines added than baseline |
| 12 | 80a61d93init: bootstrap LLM Wiki project | cmblir | 4mo ago | +1.9K -0 | 94thElevated | more lines added than baseline |
| 13 | e2164ef4feat(mcp): add standalone MCP server exposing 14 wiki tools | cmblir | 4mo ago | +823 -0 | 91thElevated | more lines added than baseline |
| 14 | bcd670b7feat(multi-project): MP-03 project resolver + /api/projects* endpoints | cmblir | 4mo ago | +459 -0 | 89thElevated | more lines added than baseline |
| 15 | 2b068a3adocs(plans): MP-01 multi-project architecture design + today-queue/backlog/blocked | cmblir | 4mo ago | +420 -0 | 89thElevated | more lines added than baseline |
| 16 | 2edd4a4adocs: README를 star-worthy 스타일로 재작성 (Memex 브랜딩 제안) | cmblirclaudeassisted | 4mo ago | +497 -391 | 89thElevated | more lines added than baseline |
| 17 | b9d6fcdcdocs: rewrite README in star-worthy style (Memex branding proposal) | cmblir | 4mo ago | +497 -391 | 89thElevated | more lines added than baseline |
| 18 | efb2551fdocs: rewrite README in star-worthy style (Memex branding proposal) | cmblir | 4mo ago | +497 -391 | 89thElevated | more lines added than baseline |
| 19 | e83379b2feat: B&W 디자인 + 모델 선택 + 사이드바 리사이즈 + 폴더 뷰 + 토스트 | cmblirclaudeassisted | 4mo ago | +604 -198 | 84thElevated | more lines added than baseline |
| 20 | 8408cc17feat: B&W design + model picker + sidebar resize + folder view + toasts | cmblir | 4mo ago | +604 -198 | 84thElevated | more lines added than baseline |
| 21 | 52a95b41feat: B&W design + model picker + sidebar resize + folder view + toasts | cmblir | 4mo ago | +604 -198 | 84thElevated | more lines added than baseline |
| 22 | daffdc87docs: README 전면 재작성 + 한국어 버전 추가 | cmblirclaudeassisted | 4mo ago | +314 -63 | 84thElevated | more lines added than baseline |
| 23 | 5718344cdocs: rewrite README + add Korean version | cmblir | 4mo ago | +314 -63 | 84thElevated | more lines added than baseline |
| 24 | e8bf4ab4docs: rewrite README + add Korean version | cmblir | 4mo ago | +314 -63 | 84thElevated | more lines added than baseline |
| 25 | 43e8fdc7feat(index): 적응형 인덱스 전략 (flat/hierarchical/indexed) | cmblirclaudeassisted | 4mo ago | +332 -4 | 84thElevated | more lines added than baseline |
| 26 | 9eeef52cfeat(index): adaptive index strategy (flat/hierarchical/indexed) | cmblir | 4mo ago | +332 -4 | 84thElevated | more lines added than baseline |
| 27 | 9ed9fca6feat(index): adaptive index strategy (flat/hierarchical/indexed) | cmblir | 4mo ago | +332 -4 | 84thElevated | more lines added than baseline |
| 28 | 847af514ingest: Test RLHF Paper | cmblir | 4mo ago | +281 -0 | 84thElevated | more lines added than baseline |
| 29 | 3b4b569eingest: Test RLHF Paper | cmblir | 4mo ago | +281 -0 | 84thElevated | more lines added than baseline |
| 30 | de4e1f00feat(multi-project): MP-07 full scoping of write + Claude-invoking endpoints | cmblir | 4mo ago | +308 -190 | 79thElevated | more lines added than baseline |
| 31 | e7a62171ingest: 제목: 안드레이 카파시(Andrej Karpathy)의 'nanochat' — 100달러로 만드는 나만의 ChatGPT | cmblir | 4mo ago | +347 -22 | 79thElevated | more lines added than baseline |
| 32 | 7e12af99ingest: title: Andrej Karpathy's 'nanochat' -- build your own ChatGPT for $100 | cmblir | 4mo ago | +347 -22 | 79thElevated | more lines added than baseline |
| 33 | 2d8cb5b9ingest: title: Andrej Karpathy's 'nanochat' -- build your own ChatGPT for $100 | cmblir | 4mo ago | +347 -22 | 79thElevated | more lines added than baseline |
| 34 | cdf6ef48feat(reflect): Reflect 오퍼레이션 추가 (위키 메타 분석) | cmblirclaudeassisted | 4mo ago | +248 -2 | 75thElevated | more lines added than baseline |
| 35 | 96ca3cfefeat(reflect): add Reflect operation (wiki meta-analysis) | cmblir | 4mo ago | +248 -2 | 75thElevated | more lines added than baseline |
| 36 | acb4bff5feat(reflect): add Reflect operation (wiki meta-analysis) | cmblir | 4mo ago | +248 -2 | 75thElevated | more lines added than baseline |
| 37 | cdedd2f4feat(ingest): diff 시각화 패널 + reasoning + ingest report | cmblirclaudeassisted | 4mo ago | +210 -11 | 75thElevated | more lines added than baseline |
| 38 | 2d568790feat(ingest): diff visualization panel + reasoning + ingest report | cmblir | 4mo ago | +210 -11 | 75thElevated | more lines added than baseline |
| 39 | 7baaee08feat(ingest): diff visualization panel + reasoning + ingest report | cmblir | 4mo ago | +210 -11 | 75thElevated | more lines added than baseline |
| 40 | 6a568a90ingest: 제목: GPT-1 (Improving Language Understanding by Generative Pre-Training, 2018) | cmblir | 4mo ago | +302 -9 | 70thElevated | more lines added than baseline |
| 41 | ddd18753ingest: title: GPT-1 (Improving Language Understanding by Generative Pre-Training, 2018) | cmblir | 4mo ago | +302 -9 | 70thElevated | more lines added than baseline |
| 42 | 05739b5bingest: title: GPT-1 (Improving Language Understanding by Generative Pre-Training, 2018) | cmblir | 4mo ago | +302 -9 | 70thElevated | more lines added than baseline |
| 43 | 2109120bdocs(schema): CLAUDE.md 전면 보강 + lint 프롬프트 체크리스트 반영 | cmblirclaudeassisted | 4mo ago | +256 -65 | 70thElevated | more lines added than baseline |
| 44 | 329a1787docs(schema): expand CLAUDE.md + align lint prompt with checklist | cmblir | 4mo ago | +256 -65 | 70thElevated | more lines added than baseline |
| 45 | 3dfe550cdocs(schema): expand CLAUDE.md + align lint prompt with checklist | cmblir | 4mo ago | +256 -65 | 70thElevated | more lines added than baseline |
| 46 | a6da9c5aingest: Test Diffusion Models | cmblir | 4mo ago | +251 -4 | 70thElevated | more lines added than baseline |
| 47 | 4f27ba15ingest: Test Diffusion Models | cmblir | 4mo ago | +251 -4 | 70thElevated | more lines added than baseline |
| 48 | 64bc0977ingest: Test Diffusion Models | cmblir | 4mo ago | +251 -4 | 70thElevated | more lines added than baseline |
| 49 | bc37d394fix(timeout): Claude CLI timeout 600s로 증가 + 환경변수 + 진단 엔드포인트 | cmblirclaudeassisted | 4mo ago | +224 -14 | 65thTypical | more lines added than baseline |
| 50 | c4b89a4efix(timeout): raise CLI timeout to 600s + env override + diagnostic endpoint | cmblir | 4mo ago | +224 -14 | 65thTypical | more lines added than baseline |
Two views of the same model: where the cuts fall, and what commit shape lands you above them.
Every scored commit, binned on the raw 0 to 10 score rather than the percentile. Percentile ranks are uniform by construction, so that axis has no shape to draw. The dashed lines are the tercile cuts behind each row's priority pill.
The 146 most recent commits, on their own recency sample rather than the feed above: that defaults to risk-sorted, so reusing it would plot only the top tercile and call it the spread. Big and scattered is what the model penalises. Click a dot to open it.
ashrafabulsoud/Memex has 148 commits in its history from 2 contributors, the first of them Apr 22, 2026. In the last 90 days 8 files were touched, 87 times in total, most often server.py. Every commit is scored for change risk from its size, spread and the history of the files it touches.
