OpenCodeReview
AI code review CLI that runs locally, sends code to a configured LLM endpoint, and checks Git diffs or whole files. Apache 2.0 licensed.
OpenCodeReview is a locally installed AI code review CLI for developers who want findings tied to specific code lines. It sends code to a configured LLM endpoint through OpenAI- or Anthropic-compatible APIs. The local CLI handles the review process; model inference depends on that endpoint.
It can review Git diffs or examine complete files when you're auditing an unfamiliar repository or working without meaningful changes to compare. Its agent can read full files, search the repository and inspect related changes, giving it context beyond the lines in a diff.
The review combines model investigation with fixed program logic for file selection, rule matching and comment placement. Related files can share a review context, while separate groups can run concurrently. This approach constrains parts of the review that a general-purpose coding agent would otherwise decide through prompts.
Built-in rules cover null pointer errors, thread safety, XSS and SQL injection across multiple languages. Custom rules can target particular paths and file characteristics. The output consists of structured comments attached to code locations.
The project reports fewer false alarms but lower defect recall than a general-purpose agent in its benchmark using the same model. That tradeoff matters if you need broad defect coverage: a quieter review can still miss problems. The CLI is open source under Apache 2.0.
