alibaba/open-code-review：README 来源编辑指南

alibaba/open-code-review 的 README 将项目描述为"Fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI &"。本文只整理仓库能直接核验的内容，不把星标、Fork 或宣传语当成质量证明。README 在"What is Open Code Review?"下的说明是：Open Code Review is an AI-powered code review CLI tool. It originated as Alibaba Group's internal official AI code review assistant , over the past two years, it has served tens of thousands of developers and identified millions of code。这给出的首先是项目边界，而不是已经完成的生产验证。

从 README 的"The Problem with General-Purpose Agents"和相关条目看，读者可以先判断它是否解决自己的具体问题：Position drift , Reported issues frequently don't match the actual code location, with line numbers or file references drifting off target.。如果你的目标与这段说明不一致，就不应仅凭项目热度采用它。这里保留原项目名、命令和组件名，方便回到一手来源核对。 README 还列出了另一条可核对的信息：Incomplete coverage , On larger changesets, agents tend to "cut corners," selectively reviewing only some files and missing others.。这类原文条目可以帮助读者设计试运行步骤，但不能代替自己的环境测试。

README 把工作方式分散写在"Benchmark"等段落中。可确认的线索包括：> Compared to general-purpose agents (Claude Code), Open Code Review achieves significantly higher Precision and F1 with the same underlying model, while consuming only 1/9 of the tokens and completing reviews faster.。这篇整理没有把未写出的架构、性能或安全边界补成结论；真正的运行链仍应结合仓库目录、配置文件和版本标签检查。

第一次安装应从 README 给出的入口开始。当前可复核的命令是： npm install -g @alibaba-group/open-code-review 如果仓库没有提供命令，本文不会替它编造安装步骤，而是建议先打开 README 的"Benchmark"部分，确认系统依赖、默认端口和首次初始化动作。

日常使用的细节取决于项目实际文档。README 的"Benchmark"段落提到：A real-world code review benchmark built from 50 popular open-source repositories, 200 real Pull Requests, and 10 programming languages , cross-validated by 80+ senior engineers (1,505 annotated ground-truth issues).。对于配置文件、环境变量、权限和数据目录，当前稿只记录来源明确的部分；未写明的默认值必须在测试环境中验证，并保留可回滚的配置副本。 同一部分还提到：Unstable quality , Natural-language-driven Skills are hard to debug, and review quality fluctuates significantly with minor prompt variations.。