feat(genkit-tools/mcp): MCP fix for Antigravity

[ssbushi]

• Accept projectRoot, if in AGY

Checklist (if applicable):

• PR title is following https://www.conventionalcommits.org/en/v1.0.0/
• Tested (manually)

[gemini-code-assist]

Summary of Changes

Hello @ssbushi, 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 focuses on enhancing the robustness and compatibility of Genkit tools, particularly within the Antigravity environment. It introduces a more centralized and context-aware approach to managing the Genkit runtime, ensuring that tools operate correctly by explicitly handling project root paths and adapting their behavior based on the execution environment. This also includes hiding certain runtime tools in Antigravity to prevent known issues, streamlining the user experience.

Highlights

• Antigravity Environment Adaptations: Implemented specific workarounds for the Antigravity environment, including requiring projectRoot as an explicit input for tools and conditionally disabling runtime management tools where Antigravity has known compatibility issues.
• Refactored Runtime Manager: The McpRuntimeManager has been refactored into a static class, ensuring a single, consistent runtime manager instance across the MCP server. It now intelligently handles changes in the projectRoot by stopping and restarting the underlying runtime when necessary.
• Dynamic Tool Input Schemas: Tool input schemas (list_flows, run_flow, start_runtime, kill_runtime, restart_runtime, get_trace) now dynamically include a projectRoot field when operating within an Antigravity environment, ensuring proper context for operations.
• Improved Process Execution Context: The startDevProcessManager now explicitly sets the current working directory (cwd) for spawned processes, enhancing the reliability and correctness of runtime process execution.
• User-Friendly Error Messages: Updated error messages related to missing Genkit runtimes to provide clearer guidance, suggesting the use of the start_runtime MCP tool or the genkit start CLI command.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 workaround for the Antigravity environment by allowing projectRoot to be passed with tool calls. This is achieved by updating tool schemas and handlers to accept projectRoot. A significant part of this change is the refactoring of McpRuntimeManager into a static class that acts as a singleton, managing a single runtime manager instance at a time. The changes look good and address the stated goal. I've added a few comments regarding code cleanup, type safety, and architectural considerations for the McpRuntimeManager refactoring.

[ssbushi]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces support for the Antigravity IDE environment within the MCP server, which is a significant feature enhancement. The changes are well-implemented, including new CLI options for IDE and timeout, and a thoughtful refactoring of tool definitions to handle projectRoot dynamically. The introduction of mcp/utils.ts to centralize common logic is a great improvement for maintainability. The McpRuntimeManager has also been nicely updated to manage runtimes on a per-project basis. Overall, this is a solid contribution. I have one suggestion to fully align the implementation with the goal of hiding runtime tools in the Antigravity environment, as mentioned in the PR description.

[ssbushi]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces support for explicitly specifying a projectRoot for MCP tools, which is needed for integration with Google Antigravity. It also adds a configurable timeout for runtime startup and allows specifying a path for debug logs. The changes are well-structured, introducing new options at the CLI level and propagating them down to the relevant tools and managers. A new utils.ts file centralizes common logic, and the McpRuntimeManager is refactored to handle different project roots.

My review includes a few suggestions to improve maintainability and consistency. I've pointed out some code duplication in tool definitions that could be refactored using a higher-order function, similar to a pattern already used in the PR. I've also suggested improvements to error handling in a couple of places to make it more informative and consistent across the tools.