feat(core): enable disableLLMCorrection by default

[SandyTao520]

TLDR

Enables the disableLLMCorrection setting by default. This change prevents the model from attempting to self-correct tool calls (like edit) when exact string matches aren't found, resulting in immediate failures instead of potentially incorrect auto-corrections.

Dive Deeper

By making this the default behavior, we prioritize precision and user control. It ensures that when a tool call fails, the user is notified immediately rather than having the LLM attempt a best-guess correction which might introduce subtle errors. This affects tools like edit and write_file that rely on editCorrector.ts.

Reviewer Test Plan

1. Run an edit command with a string that doesn't exist in the target file.
2. Observe that it fails immediately with an edit_no_occurrence_found error.
3. Verify that the model does not attempt a self-correction turn (visible in debug mode or by the lack of 'Gemini is correcting...' rationale).

Testing Matrix

	🍏	🪟	🐧
npm run	✅	❓	❓
npx	✅	❓	❓
Docker	❓	❓	❓
Podman	❓	-	-
Seatbelt	❓	-	-

Linked issues / bugs

None

[gemini-code-assist]

Summary of Changes

Hello @SandyTao520, 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 significant change to the default behavior of LLM-based error correction within the system's tools. By enabling disableLLMCorrection by default, the system will now prioritize immediate failure and explicit user notification when tool calls, particularly those involving string matching in edit and write_file operations, do not find exact matches. This shift aims to prevent potentially incorrect or subtle auto-corrections by the LLM, ensuring greater precision and user control over tool execution outcomes.

Highlights

• Default Behavior Change: The disableLLMCorrection setting is now enabled by default, preventing the LLM from attempting self-correction for tool calls like edit and write_file.
• Prioritizing Precision: This change ensures that tools fail immediately if exact string matches are not found, providing clearer and more precise feedback to the user instead of potentially incorrect auto-corrections.
• Documentation and Schema Updates: All relevant documentation and configuration schemas have been updated to reflect the new default value for disableLLMCorrection.
• Test Adjustments: Existing tests for edit and write_file tools have been updated to account for the new default, with explicit overrides where LLM correction is still being tested.

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. ↩

[NTaylorMullen]

File a follow up to remove this feature entirely if you could!

[gemini-code-assist]

Code Review

This pull request successfully implements the feature to enable disableLLMCorrection by default across the application. The changes are consistently applied in the core configuration, settings schema, and relevant documentation. Test files have been appropriately updated to reflect this new default behavior, with specific tests explicitly re-enabling LLM correction to ensure continued coverage of that functionality. The changes align well with the stated objective of prioritizing precision and user control by preventing automatic LLM self-correction for tool calls.

[github-actions]

Size Change: -5 B (0%)

Total Size: 23.1 MB

ℹ️ View Unchanged

Filename	Size	Change
./bundle/gemini.js	23.1 MB	-5 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}