fix: prevent tools discovery error for prompt-only MCP servers

[vasayxtx]

TLDR

This fixes the "-32601: tools not supported" error that occurs when connecting to MCP servers that only provide prompts (no tools).

Dive Deeper

Similar to how discoverPrompts() checks for prompt support:
if (mcpClient.getServerCapabilities()?.prompts == null) return [];

The discoverTools() function should check for tool support before attempting to call mcpToTool(), and silently return an empty array if tools are not supported.

Reviewer Test Plan

Connect to an MPC server which doesn't implement the tools/list handler, you will no longer see an error on startup (but would previously).

Testing Matrix

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

Linked issues / bugs

Fixes #10365

[gemini-code-assist]

Summary of Changes

Hello @vasayxtx, 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 an issue where the MCP client would encounter an error when attempting to discover tools on servers that are configured to only support prompts. By introducing a pre-check for server capabilities within the discoverTools() function, the client now intelligently skips tool discovery if the server explicitly indicates no tool support, thereby preventing errors and improving the overall stability and compatibility with diverse MCP server configurations.

Highlights

• Error Prevention: Implemented a check in the discoverTools() function to prevent a "-32601: tools not supported" error when connecting to MCP servers that only provide prompt capabilities and do not support tools.
• Robustness Improvement: Ensures that the client gracefully handles servers lacking tool support by silently returning an empty array for tool discovery, mirroring the existing behavior for prompt discovery.
• Enhanced Test Coverage: Added new unit tests to mcp-client.test.ts to explicitly verify the correct behavior of tool discovery, covering scenarios where servers either support or do not support tools.

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 correctly fixes an error that occurred when connecting to prompt-only MCP servers. The change adds a check for tool support in discoverTools before proceeding, which is consistent with how discoverPrompts works. The accompanying tests are well-written, covering both scenarios where a server supports tools and where it does not. The changes are sound and address the issue effectively.

[vasayxtx]

The CI failures appear to be flaky tests - both fail at interactive mode startup, which is unrelated to my MCP capability checking changes. @scidomino, could you re-run the failed jobs? Thanks!

[scidomino]

There was a flakey test fix that was checked in recently that you didn't have. I updated your branch and am running them again.