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
3,296 commits sit in this repo's top risk tercile, which is 33% of the 10,000scored. 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 7.1 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.
Began fix-led, now leaning chore.
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 | Risk | Top driver |
|---|---|---|---|---|---|---|
| 1 | 0d2cea74docs: streamline LangSmith teasing (#30302) | Oskar Stark | 1y ago | +22.3K -22.4K | 100%Elevated | more lines added than baseline |
| 2 | d172984cinfra: migrate to uv (#29566) | ccurme | 1y ago | +56.3K -62.3K | 100%Elevated | more lines added than baseline |
| 3 | 21d14549docs: v0.2 docs in master (#21438) | Erick Friis | 2y ago | +22.0K -53.4K | 100%Elevated | more lines added than baseline |
| 4 | 49f3cc0ftemplates: bump lockfile deps (#19001) | Erick Friis | 2y ago | +36.1K -34.7K | 100%Elevated | more lines added than baseline |
| 5 | 223e5effAdd JSON representation of runnable graph to serialized representation (#17745) | Nuno Campos | 2y ago | +51.5K -1.2K | 100%Elevated | more lines added than baseline |
| 6 | 64785822templates: bump (#17074) | Erick Friis | 2y ago | +43.8K -41.5K | 100%Elevated | more lines added than baseline |
| 7 | 69a8a266templates: fix deps (#15439) | Erick Friis | 2y ago | +59.6K -46.4K | 100%Elevated | more lines added than baseline |
| 8 | 67eca93dchore(model-profiles): refresh model profile data (#37466) | langchain-model-profile-bot[bot] | 2mo ago | +5.4K -1.9K | 100%Elevated | more lines added than baseline |
| 9 | 9ce72ebafeat(core): add content-block-centric streaming (v2) (#36834) | Nick Hollon | 3mo ago | +8.4K -96 | 100%Elevated | more lines added than baseline |
| 10 | f9fd7be6feat(openrouter): add `langchain-openrouter` provider package (#35211) | Mason Daugherty | 5mo ago | +9.0K -4 | 100%Elevated | more lines added than baseline |
| 11 | 637bb1cbfeat: refactor tests coverage (#33927) | Sydney Runkle | 9mo ago | +5.3K -4.9K | 100%Elevated | more lines added than baseline |
| 12 | 493be259feat(core): mint langchain-model-profiles and add profile property to `BaseChatModel` (#33728) | ccurme | 9mo ago | +17.3K -0 | 100%Elevated | more lines added than baseline |
| 13 | 6fc21afbstyle: `.. code-block::` admonition translations (#33400) | Mason Daugherty | 10mo ago | +10.4K -11.3K | 100%Elevated | more lines added than baseline |
| 14 | eaa6dccerelease: v1.0.0 (#32567) | Mason Daugherty | 10mo ago | +23.8K -17.6K | 100%Elevated | more lines added than baseline |
| 15 | 7ddc798ffix(openai): pin upper bound to prevent Pydantic 2.7.0 issues (#33038) | Mason Daugherty | 10mo ago | +7.1K -5.6K | 100%Elevated | more lines added than baseline |
| 16 | 18db07c2feat(langchain): revamped `create_react_agent` (#32705) | Sydney Runkle | 11mo ago | +8.1K -69 | 100%Elevated | more lines added than baseline |
| 17 | 4d9eefecfix: bump lockfiles (#31923) | Mason Daugherty | 1y ago | +5.7K -5.7K | 100%Elevated | more lines added than baseline |
| 18 | f636c833community: Cassandra Vector Store: modernize implementation (#27253) | Eric Pinzur | 1y ago | +4.1K -667 | 100%Elevated | more lines added than baseline |
| 19 | f9ef688bdocs: upgrade to docusaurus v3 (#26803) | Erick Friis | 1y ago | +11.2K -14.4K | 100%Elevated | more lines added than baseline |
| 20 | c2a3021bmultiple: pydantic 2 compatibility, v0.3 (#26443) | Erick Friis | 1y ago | +38.8K -30.9K | 100%Elevated | more lines added than baseline |
| 21 | a72fddbf[docs]: vector store integration pages (#24858) | Isaac Francisco | 2y ago | +5.3K -4.1K | 100%Elevated | more lines added than baseline |
| 22 | 838464deollama: init package (#23615) | Isaac Francisco | 2y ago | +8.7K -510 | 100%Elevated | more lines added than baseline |
| 23 | 404d92demilvus: New langchain_milvus package and new milvus features (#21077) | ChengZi | 2y ago | +7.3K -28 | 100%Elevated | more lines added than baseline |
| 24 | edd68e4aqdrant: init package (#21146) | Anush | 2y ago | +5.7K -47 | 100%Elevated | more lines added than baseline |
| 25 | f92006demultiple: langchain 0.2 in master (#21191) | Eugene Yurtsev | 2y ago | +7.6K -5.9K | 100%Elevated | more lines added than baseline |
| 26 | 0e5bf16dlangchain[patch]: Migrate document loaders to use optional langchain community imports (#21095) | Eugene Yurtsev | 2y ago | +4.4K -721 | 100%Elevated | more lines added than baseline |
| 27 | cc6191cbcommunity[minor]: Add support for Upstash Vector (#20824) | Cahid Arda Öz | 2y ago | +11.8K -4 | 100%Elevated | more lines added than baseline |
| 28 | 2fa0ff1acli[minor]: update code to generate migrations from langchain to community (#20946) | Eugene Yurtsev | 2y ago | +5.4K -2.4K | 100%Elevated | more lines added than baseline |
| 29 | 2ae6dcdfcore: Assign missing message ids in BaseChatModel (#19863) | Nuno Campos | 2y ago | +8.1K -56.7K | 100%Elevated | more lines added than baseline |
| 30 | 01759064templates: add RAG template for Intel Xeon Scalable Processors (#18424) | lvliang-intel | 2y ago | +6.0K -0 | 100%Elevated | more lines added than baseline |
| 31 | 781aee00community, langchain, infra: revert store extended test deps outside of poetry (#19153) | Erick Friis | 2y ago | +11.1K -742 | 100%Elevated | more lines added than baseline |
| 32 | 328a498adocs[minor]: Add thumbs up/down to all docs pages (#18526) | Brace Sproul | 2y ago | +13.5K -1 | 100%Elevated | more lines added than baseline |
| 33 | 76eb5530docs: add documentation for Google Cloud database integrations (#18225) | Averi Kitsch | 2y ago | +9.8K -153 | 100%Elevated | more lines added than baseline |
| 34 | 50ba3c68community[minor]: add Kinetica LLM wrapper (#17879) | Chad Juliano | 2y ago | +6.1K -1 | 100%Elevated | more lines added than baseline |
| 35 | 12d2b2ebdocs[minor]: LCEL rewrite of chatbot use-case (#16414) | Jacob Lee | 2y ago | +3.0K -747 | 100%Elevated | more lines added than baseline |
| 36 | 08be477ctemplates: 0.1 bump (#15648) | Erick Friis | 2y ago | +10.0K -7.8K | 100%Elevated | more lines added than baseline |
| 37 | ebc75c5copenai[minor]: implement langchain-openai package (#15503) | Erick Friis | 2y ago | +6.0K -389 | 100%Elevated | more lines added than baseline |
| 38 | f20c56db[documentation] documentation revamp (#15281) | Harrison Chase | 2y ago | +10.3K -15.3K | 100%Elevated | more lines added than baseline |
| 39 | 6c2b70e6feat(openai): add ChatGPT OAuth-backed `ChatOpenAICodex` chat model (#37569) | open-swe[bot] | 2mo ago | +4.3K -7 | 99%Elevated | more lines added than baseline |
| 40 | 1de100f2chore(infra): bump mypy to 2.1 and unify type-check config across the monorepo (#36470) | Christophe Bornet | 2mo ago | +3.8K -1.5K | 99%Elevated | more lines added than baseline |
| 41 | 4000c223feat(openai): prevent silent streaming hangs in `ChatOpenAI` (#36949) | Asamu David | 3mo ago | +2.0K -49 | 99%Elevated | more lines added than baseline |
| 42 | 2dc13963chore(langchain): update deps (#33951) | Sydney Runkle | 9mo ago | +1.9K -1.6K | 99%Elevated | more lines added than baseline |
| 43 | a022e3c1feat(langchain_v1): Add ShellToolMiddleware and ClaudeBashToolMiddleware (#33527) | Nuno Campos | 9mo ago | +2.3K -477 | 99%Elevated | more lines added than baseline |
| 44 | 34de8ec1feat(anthropic): add more anthropic middleware (#33510) | Eugene Yurtsev | 9mo ago | +2.5K -169 | 99%Elevated | more lines added than baseline |
| 45 | 1bf29da0feat(langchain_v1): add on_tool_call middleware hook (#33329) | Eugene Yurtsev | 10mo ago | +2.6K -43 | 99%Elevated | more lines added than baseline |
| 46 | 2c3fec01feat(langchain_v1): on_model_call middleware (#33328) | Eugene Yurtsev | 10mo ago | +2.3K -195 | 99%Elevated | more lines added than baseline |
| 47 | af07949dfix(docs): Redirects (#33190) | Lauren Hirata Singh | 10mo ago | +3.2K -1 | 99%Elevated | more lines added than baseline |
| 48 | 781db9d8chore: update `pyproject.toml` files, remove codespell (#33028) | Mason Daugherty | 10mo ago | +6.4K -4.9K | 99%Elevated | more lines added than baseline |
| 49 | bd765753release(chroma): 0.2.6 (#32901) | Mason Daugherty | 11mo ago | +2.2K -1.7K | 99%Elevated | more lines added than baseline |
| 50 | 6e2f46d0feat(langchain): middleware support in `create_agent` (#32828) | Sydney Runkle | 11mo ago | +2.7K -293 | 99%Elevated | 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 200 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.
langchain-ai/langchain has 16,571 commits in its history from 3,600 contributors, the first of them Oct 24, 2022. In the last 90 days 454 files were touched, 1,140 times in total, most often _profiles.py. Every commit is scored for change risk from its size, spread and the history of the files it touches.
