PR Pilot Review

Intelligent Multi-Model Consensus AI PR Reviewer — Automated code reviews with consensus from configurable LLM providers like Gemini and Groq.

Overview

PR Pilot Review is a GitHub Action that performs intelligent, consensus-based code reviews on pull requests using configurable LLM providers. Unlike single-model reviewers, it runs three independent AI reviewers in parallel and employs a judge model to reach consensus, ensuring more thorough, balanced, and objective code quality feedback.

Key Features

• Multi-Model Consensus: Utilizes 3 independent reviewers and 1 judge for balanced decisions.
• Parallel Execution: All reviewers run simultaneously, typically completing within seconds.
• Inline Comments: Provides specific code suggestions directly on exact lines in the pull request diff.
• Idempotent Operations: Skips previously reviewed commits to avoid redundant processing.
• Smart Filtering: Automatically ignores build artifacts, binary files, and dependencies to focus on critical changes.
• Highly Configurable: Allows selection of reviewer models to optimize for cost, speed, or quality.
• Safe Defaults: Employs continue-on-error behavior to ensure the action never blocks pull request workflows.
• Production Ready: Built with robust error handling and graceful degradation mechanisms.

Documentation

Comprehensive documentation has been organized into specific guides:

• Quick Start Guide: Step-by-step instructions for integrating the action into your repository, configuring API keys, and setting up workflows.
• API Reference: Detailed information on action inputs, outputs, model presets, and configuration options.
• Architecture Overview: High-level explanation of the system's consensus algorithm, decision logic, and internal structure.
• Development Guide: Instructions for local environment setup, testing, and building the project.
• Contributing: Guidelines for opening issues, submitting pull requests, and adhering to code standards.

For in-depth architectural decisions, please refer to our Architecture Decision Records (ADRs) located in docs/adr/.

API Keys & Billing

This action requires an API key from a supported LLM provider (e.g., Gemini or Groq). Please refer to the API Reference for configuration details. Be aware of your provider's rate limits, free tier quotas, and potential costs associated with the models you select.

Troubleshooting

If you encounter issues such as the action not running, zero findings being reported, or quota exceeded errors, please consult the troubleshooting section in the Quick Start Guide. Enabling debug mode (debug: "true") can provide detailed logs of the consensus process.

Security

• No Code Execution: The action only reads pull request diffs and does not execute the source code.
• API Key Security: Keys are securely passed via GitHub Secrets and are not logged.
• Audit Trail: All reviews are submitted natively as GitHub PR comments, providing a clear history.
• Fail-Safe Design: Uses continue-on-error by default to prevent blocking critical CI/CD pipelines upon failure.

License

This project is licensed under the MIT License — See the LICENSE file for details.

---

Developed by Bishal Prasad

For further questions or discussions, please open an issue or refer to our Discussions page.