open-code-review

What it does

open-code-review is an intelligent code review assistant that combines deterministic analysis pipelines with LLM agent capabilities. It performs repository-level code analysis, generating precise line-level comments and detecting security vulnerabilities, concurrency issues, and code quality problems across multiple programming languages.

How to set up

Clone the repository from GitHub. Install dependencies based on your environment. Configure API credentials for OpenAI or Anthropic. Run the tool via CLI against your codebase or integrate it as part of your CI/CD pipeline. Supports dynamic concurrency configuration for optimal performance on large repositories.

Key features

• Deterministic pipelines combined with LLM agent reasoning for accurate code analysis
• Precise line-level comments with context-aware suggestions
• 10+ language-specific rules including NPE, thread-safety, XSS, and SQL injection detection
• OpenAI and Anthropic API compatibility
• Smart context compression for handling large codebases
• Dynamic concurrency management for efficient processing
• Repository-level context awareness for holistic code understanding