Open sourceGo · ≈49k stars · Apache-2.0
Open Code Review
Alibaba's hybrid deterministic-plus-agent code review CLI.
Open Code Review is an AI code review CLI open-sourced by Alibaba, where it served as the internal official review assistant for two years. It reads Git diffs and passes changed files to a configurable OpenAI- or Anthropic-compatible LLM via a tool-using agent that can read files and search the codebase, producing line-level comments. Its design pairs deterministic pipelines (hard constraints on coverage and positions) with the agent, plus a built-in multi-language ruleset. Its README reports higher precision and F1 than Claude Code on a 200-PR benchmark at about 1/9 the tokens, with lower recall.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- DocsComes from Alibaba's internal AI review assistant, which the README says served tens of thousands of developers over two years.[1]
- DocsIts own benchmark (200 PRs, 50 repos, 1,505 annotated issues) reports higher precision and F1 than Claude Code with the same model, using about 1/9 of the tokens.[1]
- User reportsA user reports good experience with it, and HN commenters see the curated per-language rule files as the main value.[1][2]
- MetadataApache-2.0, a single CLI, and frequent releases (v1.13.3, Oct 2026).[1]
Cons
- DocsLower recall than general-purpose agents by design: the README says it trades recall for precision.[1]
- User reportsThe rule files were originally written in Chinese; English came later, after a community Google-Translate fork.[1][2]
- User reportsThe headline benchmark is its own. The only third-party comparison found (run by a rival tool's author) completed just a partial run, 4 of 20 PRs.[1]
- MetadataYoung as a public project: first published in May 2026.[1]
- Best for
- Teams that want low-noise, rule-driven AI review comments in CI at modest token cost.
- Watch out for
- It will miss some issues by design; pair it with a broader reviewer or a human on risky changes.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
People
Engineer
Creator & Project Lead, Open Code Review (Alibaba)
Comparisons & reviews
In research (1)
- CIVER: Causal Intervention Verification for Reliable Code ReviewVerdict: Incremental
Oct 26, 2026 · OpenCodeReview is cited as a deployed LLM review agent whose findings CIVER-style verification could filter.
Neighbours on the workbench
MDX · ≈307k stars · MIT
obra/superpowers
An agentic skills framework and software methodology that works.
Maintained by Jesse Vincent (obra)
Shell · ≈306k stars · MIT
mattpocock/skills
Small, composable agent skills for real engineering work.
Maintained by Matt Pocock (AI Hero)
JavaScript · ≈292k stars · MIT
affaan-m/ECC
Agent harness performance optimization system: skills, instincts, memory and security for coding agents.
Maintained by Affaan Mustafa
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Maintained by · 1
- AlibabaOrganisation
Built on · 1
- Claude CodeProject
Sources
Verified Oct 29, 2026