code-review

To get unified diffs from multiple AI code review providers, you submit a structured Markdown request to the CLI, which coordinates GitHub Copilot, Claude, Codex, and Gemini to return comparable patches for your changes.

Yes, you can include uncommitted workspace changes in an AI code review patch request by adding the workspace option, allowing the configured providers to analyze and generate diffs based on your local working environment.

The best way to compare AI code review patches from different models is using a multi-provider CLI that submits a single structured request to multiple engines and returns unified diffs, enabling direct comparison of diverse insights.

Yes, the code review pipeline outputs JSON-formatted results that embed the original prompts, provider responses, and optional diff blocks, ensuring seamless integration into downstream automation workflows.

Running a complete AI code review pipeline involves providing a Markdown change description, selecting providers and models, optionally adding workspace access, and executing the three-step pipeline to obtain structured patches.

Using multiple AI providers for code review instead of a single engine enables teams to gather diverse insights, compare unified diffs from different models, and accelerate code-quality improvements by leveraging varied AI perspectives.