fix: Fix issue with MCP tools throwing an error.

[nickvander]

Problem

When an MCP tool is resolved using _gemini_schema_util.py, if it has an array it wasn't correctly resolving them and showing an error:

GenerateContentRequest.tools[0].function_declarations[1].parameters.properties[segmentation_classes].items: missing field

Root Cause

The error "parameters.properties[segmentation_classes].items: missing field" indicates _to_gemini_schema isn't correctly handling array types.

Implementation Details

Correctly handle array types by adding a default items field of {"type": "string"} to schemas with type: array but no items field.

Associated Issue (fixes)

#3082

Testing Plan

Reran the MCP server that wasn't working, and this now works. This is the MCP server that I used:

https://github.com/GoogleCloudPlatform/vertex-ai-creative-studio/blob/main/experiments/mcp-genmedia/mcp-genmedia-go/mcp-imagen-go/README.md

Verification

With this update, the MCP servers now work without issues!

[gemini-code-assist]

Summary of Changes

Hello @nickvander, 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 addresses a critical bug affecting MCP tools that utilize _gemini_schema_util.py for schema resolution. Previously, tools with array types in their schemas that omitted an explicit items field would encounter a "missing field" error, preventing their proper functionality. The change introduces a robust solution by automatically injecting a default {"type": "string"} for the items field into such array schemas, thereby ensuring correct schema interpretation and enabling MCP tools to operate without errors.

Highlights

• MCP Tool Resolution Fix: Resolved an issue where MCP tools failed to resolve correctly when their schemas contained array types without an explicit "items" field, leading to a "missing field" error.
• Schema Handling for Arrays: Implemented a fix in _gemini_schema_util.py to correctly handle array types by adding a default {"type": "string"} for the items field to schemas that are of type 'array' but lack this field.

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. ↩

[adk-bot]

Hello @nickvander, thank you for submitting this pull request!

I noticed that the pyink-check is failing. This is likely due to a code formatting issue. You can fix this by running ./autoformat.sh locally and pushing the changes.

You can find more information about the development setup and contribution workflow in our contribution guidelines.

Response from ADK Triaging Agent

[gemini-code-assist]

Code Review

This pull request addresses an issue with MCP tools that would error when processing array types in schemas that were missing an items field. The fix correctly introduces a default items field for such cases. My review includes a suggestion to refactor the implementation to eliminate code duplication, which will enhance the code's maintainability.