The best code reviews might start with deterministic rules, not the llm. alibaba's…
the best code reviews might start with deterministic rules, not the llm. alibaba's open-code-review, trending september 16, runs fixed pipelines for npe, thread-safety, xss and sql injection before the agent writes line-level comments.
the llm argues with evidence.
Context
The alibaba/open-code-review repository describes itself as deterministic pipelines plus an LLM agent for code review, with a built-in multi-language ruleset covering NPE, thread-safety, XSS and SQL injection. Its design makes file selection, file bundling and rule matching deterministic. The README reports a benchmark of 50 repositories, 200 pull requests, 10 languages and 1,505 annotated issues, with higher precision and F1 at roughly one ninth of the tokens and lower recall. Trendshift shows the repository first reaching number one on GitHub Trending on July 23, 2026 and number one on Trendshift on September 14, 2026.
The note's order, fixed pipelines before the agent writes comments, is consistent with the deterministic selection and bundling design, but the execution order was not traced in code. The benchmark is vendor-reported and not independent, and a Hacker News commenter's roughly 12 percent precision on 10 pull requests used a different benchmark and is not a matched refutation. A September 16 trending date was seen only in a search snippet and not confirmed; Trendshift's dates differ. The repository creation date, 2026-05-18, does not establish the open-sourcing date, and the star count is a later snapshot. The LLM argues with evidence is the author's take.
Related work
- the best code review this week is half deterministic rules, half llm. ↗An earlier note on the same project.
- the most interesting review tool right now runs a deterministic pipeli ↗An earlier note on the same project.
- alibaba's open code review beats claude code on its own benchmark with ↗An earlier note on the same project.
- the smartest ai code review tool this month is the one that barely use ↗An earlier note on the same project.
Watch next
- Independent evaluations on matched benchmarks.
Sources
- GitHub: alibaba/open-code-reviewgithub.com
- READMEgithub.com
- Trendshift: repository 41087trendshift.io
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 22 September 2026 at 14:19 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →