feat(plan): add experimental 'plan' approval mode

[jerop]

Introduces a new 'plan' approval mode, designed to support a future read-only agent state where write operations are automatically blocked or restricted. This foundation allows the system to distinguish between normal interactive modes and this restricted planning state.

The new mode is currently gated behind the experimental.plan setting to prevent premature usage while the underlying policy rules are being developed. When enabled, it allows users to start the CLI in this specific mode via the --approval-mode=plan flag.

Closes #16618
Closes #16619

[gemini-code-assist]

Summary of Changes

Hello @jerop, 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 lays the foundational groundwork for a future read-only agent state by introducing an experimental 'plan' approval mode. This mode, accessible via a new CLI flag and controlled by an experimental setting, will allow the system to operate in a restricted planning state, distinguishing it from normal interactive modes. The changes include updates to configuration, documentation, and robust validation to ensure controlled access to this new feature.

Highlights

• New 'plan' Approval Mode: Introduced an experimental 'plan' approval mode, designed to support a future read-only agent state where write operations are automatically blocked or restricted.
• Gated by Experimental Setting: The new 'plan' mode is gated behind the experimental.plan setting, preventing premature usage while underlying policy rules are developed.
• CLI Flag Support: Users can now start the CLI in this specific mode using the --approval-mode=plan flag, provided the experimental setting is enabled.
• Documentation Update: The docs/get-started/configuration.md file has been updated to include the new plan approval mode and its experimental status.
• Validation and Test Coverage: Added validation logic to ensure the 'plan' mode is only accessible when experimental.plan is enabled, along with comprehensive test cases to cover these scenarios.

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 a new experimental 'plan' approval mode, which is intended to support a future read-only agent state. The changes correctly add the new mode and gate it behind an experimental flag. My review focuses on ensuring the new mode is integrated correctly. I've found a couple of issues in packages/cli/src/config/config.ts: an inconsistent error message and a missing case in a switch statement that leads to incorrect behavior for the new 'plan' mode in non-interactive sessions. Addressing these will improve correctness and user experience.

[github-actions]

Size Change: +463 B (0%)

Total Size: 23.1 MB

ℹ️ View Unchanged

Filename	Size	Change
./bundle/gemini.js	23.1 MB	+463 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}

[jacob314]

I'll fast follow with the fix