fix(core): prevent infinite loops with hard cap on consecutive same-name tool calls

[noxymon]

Summary

This PR implements a hard cap for consecutive calls to the same tool name to prevent infinite loops, especially in "thinking" tools where arguments (like thought numbers) change each time. It also enhances the loop detection event to carry more details to the UI.

Details

• Added CONSECUTIVE_SAME_TOOL_NAME_THRESHOLD (50) and THINKING_TOOL_LOOP_THRESHOLD (15) to LoopDetectionService.
• Enhanced loop detection to track consecutive calls to the same tool name even if arguments differ.
• Updated LoopDetected event to carry specific loop details ( ype, count, detail).
• Improved UI alerting in useGeminiStream to display detailed information about the detected loop.
• Added comprehensive unit tests for the new loop detection logic.

Related Issues

Fixes #25671

How to Validate

1. Run
   pm test -w @google/gemini-cli-core -- loopDetectionService.test.ts to verify the logic.
2. Trigger a loop in the CLI with a tool that changes arguments (e.g. sequentialthinking) and verify it stops after 15 calls with a detailed message.

Pre-Merge Checklist

• Updated relevant documentation and README (if needed)
• Added/updated tests (if needed)
• Noted breaking changes (if any)
• Validated on required platforms/methods:
  • MacOS
    • npm run
    • npx
    • Docker
    • Podman
    • Seatbelt
  • Windows
    • npm run
    • npx
    • Docker
  • Linux
    • npm run
    • npx
    • Docker

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[gemini-code-assist]

Summary of Changes

Hello, 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 significantly enhances the system's ability to detect and prevent infinite loops during tool calls by introducing new thresholds and tracking mechanisms. It specifically addresses cases where tool arguments might change, which previously bypassed simpler loop detection. The changes also improve the user experience by providing more detailed information when a loop is identified, allowing for better debugging and understanding of model behavior.

Highlights

• Enhanced Loop Detection: Implemented a hard cap for consecutive calls to the same tool name, even if arguments differ, to prevent infinite loops, particularly for 'thinking' tools.
• Configurable Thresholds: Introduced CONSECUTIVE_SAME_TOOL_NAME_THRESHOLD (50) and THINKING_TOOL_LOOP_THRESHOLD (15) to LoopDetectionService for more granular control over loop detection.
• Detailed Loop Information: The LoopDetected event now carries specific details about the detected loop (type, count, detail, confirmedByModel) to provide richer context.
• Improved UI Feedback: The UI alerting in useGeminiStream was updated to display the detailed information from the LoopDetected event, enhancing user understanding.
• Comprehensive Unit Tests: Added extensive unit tests to cover the new loop detection logic, including scenarios with varying arguments for consecutive tool calls.

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.

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 enhances loop detection by tracking consecutive calls to the same tool name even when arguments differ, which is particularly useful for identifying loops in 'thinking' tools. It updates the UI to display specific loop details and adds corresponding unit tests. Feedback suggests introducing a distinct telemetry event type for same-name tool calls to distinguish them from truly identical repetitions, ensuring more accurate data analysis.