fix(patch): cherry-pick 2995af6 to release/v0.21.0-preview.5-pr-15131 to patch version v0.21.0-preview.5 and create version 0.21.0-preview.6

[gemini-cli-robot]

This PR automatically cherry-picks commit 2995af6 to patch version v0.21.0-preview.5 in the preview release to create version 0.21.0-preview.6.

[gemini-code-assist]

Summary of Changes

Hello @gemini-cli-robot, 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 is an automated cherry-pick designed to update a preview release, specifically patching version v0.21.0-preview.5 to create v0.21.0-preview.6. The core change introduces a mechanism within the a2a-server to conditionally utilize a preview version of the Gemini model, enabling developers to test experimental features when a specific setting is active.

Highlights

• Conditional Model Selection: The a2a-server configuration now dynamically selects between PREVIEW_GEMINI_MODEL and DEFAULT_GEMINI_MODEL based on the settings.general?.previewFeatures flag. This allows for easy toggling of preview model usage.
• Import PREVIEW_GEMINI_MODEL: The PREVIEW_GEMINI_MODEL constant has been imported from @google/gemini-cli-core to support the new conditional model selection logic.

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. ↩

[github-actions]

Size Change: -2 B (0%)

Total Size: 21.6 MB

ℹ️ View Unchanged

Filename	Size	Change
./bundle/gemini.js	21.6 MB	-2 B (0%)
./bundle/sandbox-macos-permissive-closed.sb	1.03 kB	0 B
./bundle/sandbox-macos-permissive-open.sb	890 B	0 B
./bundle/sandbox-macos-permissive-proxied.sb	1.31 kB	0 B
./bundle/sandbox-macos-restrictive-closed.sb	3.29 kB	0 B
./bundle/sandbox-macos-restrictive-open.sb	3.36 kB	0 B
./bundle/sandbox-macos-restrictive-proxied.sb	3.56 kB	0 B

_{compressed-size-action}

[gemini-code-assist]

Code Review

This pull request cherry-picks a change to use a preview Gemini model when preview features are enabled. The implementation has a potential race condition where the model is selected before the previewFeatures flag is finalized, as it can be updated remotely during authentication. My review includes a comment with high severity to address this correctness issue, suggesting a refactor to determine the model after authentication is complete to ensure the correct model is used.

[gemini-code-assist]

There's a potential issue with how the model is selected. The model is determined here based on the settings.general?.previewFeatures value. However, for users authenticating with Google (USE_CCPA), the previewFeatures flag can be enabled later via experiments inside the config.refreshAuth() call.

When this happens, config.previewFeatures will become true, but the model will remain DEFAULT_GEMINI_MODEL because this logic is not re-evaluated. This can lead to an inconsistent state where preview features are enabled but the required preview model is not being used.

To resolve this, I recommend setting the model after refreshAuth has been called to ensure you are using the final value of previewFeatures. You can set the default model here, and then conditionally update it to PREVIEW_GEMINI_MODEL after the authentication block later in this function.

    model: DEFAULT_GEMINI_MODEL,