RFC: Reflection API V2 (WebSocket)

[pavelgj]

Introduces a new Reflection API (V2) based on WebSockets and JSON-RPC 2.0. This architecture reverses the connection direction of V1: the Genkit CLI (Runtime Manager) acts as the WebSocket server, and Genkit Runtimes connect to it as clients.

Related RFCs:

• RFC: [JS] Bidirectional Actions, Flows, and Models #4210

[gemini-code-assist]

Summary of Changes

Hello @pavelgj, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces a comprehensive RFC for Reflection API V2, a significant architectural upgrade for Genkit. By adopting WebSockets and JSON-RPC 2.0, the new API reverses the connection model, positioning the Genkit CLI as the server and runtimes as clients. This change fundamentally improves the system's ability to handle bidirectional actions, broadens environmental compatibility, and streamlines the process of runtime discovery, all while maintaining compatibility with the existing V1 API.

Highlights

• New Reflection API (V2): Introduces a new version of the Reflection API for Genkit, designed to address limitations of the previous V1.
• WebSocket and JSON-RPC 2.0: The new API is built upon WebSockets for full-duplex communication and uses JSON-RPC 2.0 as its protocol.
• Reversed Connection Direction: The architecture shifts from V1, with the Genkit CLI (Runtime Manager) now acting as the WebSocket server and Genkit Runtimes connecting to it as clients.
• Enhanced Capabilities: V2 enables bidirectional actions, supports a wider range of environments (e.g., web browsers), and simplifies runtime discovery compared to V1.
• Coexistence with V1: The new API is designed to coexist with the existing V1, allowing the Genkit CLI to determine which mode to use.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces an RFC for a new WebSocket-based Reflection API (V2). The design is well-structured and addresses key limitations of the V1 API. My review focuses on adding more detail to the RFC to ensure a robust and secure implementation. The main feedback points include adding a security section to address authentication and encryption, defining error handling and connection management strategies, considering a heartbeat mechanism for connection stability, and clarifying the parameters for the register method. These additions will help create a more complete and resilient design.

[xavidop]

This RFC looks solid. Compared to the current implementation, it feel s more scalable, and closer to what we would expect from a modern, agent-oriented framework.

That said, adopting this RFC introduces an important question around backwards compatibility. If this new API coexists with the previous one, we will need to support two mental models and two sets of primitives. This can increase cognitive load for users and maintainers.

It would be good to clarify early how we plan to handle this transition. For example, is the intention to formally deprecate the older API over time, or will it remain a first-class option indefinitely? Are there plans for migration paths, adapters, or compatibility layers that allow users to move from the old versions to the new one without rewriting everything?

overall, I really like this new API.