open-code-review: hybrid code review from Alibaba
Alibaba's open-source code review tool. Deterministic rules plus an LLM agent, with line-level comments. About 49,000 GitHub stars when I checked in October 2026.
open-code-review is Alibaba’s open-source code review tool. The architecture is the reason to read it. Deterministic pipelines for the bugs a rule can name, and an LLM agent for the comment a rule can’t write. Line-level comments, with rulesets for the usual suspects: null pointers, thread safety, XSS, SQL injection. It speaks OpenAI-compatible and Anthropic-compatible APIs. About 49,000 GitHub stars when I checked in October 2026.
Why it stays bookmarked
A review bot that is only a model will miss the boring bug and invent a style nit. A linter will never explain a design. The hybrid is the version I want: the pipeline finds the class of bug, the model writes the note a teammate would actually read. Alibaba says it is battle-tested at their scale. The part I can check from the outside is the shape of the system, and the shape is right.
ai-sdk-computer-use is a smaller Vercel reference for a different loop, an agent that looks at a screen and acts.