Integrates with Google Cloud's Vertex AI to access Gemini 3.1 models, enabling complex research, search grounding, URL analysis, and Python code execution.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Gemini Agent MCP Server compare the capabilities of GPT-4 and Gemini 3.1"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An advanced Model Context Protocol (MCP) server that provides a high-level agentic interface to Google's Gemini 3.1 models via Vertex AI.
Unlike a standard search tool, this server exposes a single "Agent" tool that combines real-time Google Search, deep URL analysis, and Python code execution to solve complex, multi-step research and data tasks.
Search Grounding: Uses Google Search to find up-to-the-minute information.
URL Context: Automatically fetches and parses the content of specific web pages for deep analysis.
Code Execution: Writes and executes Python code on-the-fly to perform calculations, data manipulation, or logical reasoning.
Thinking Mode: Utilizes Gemini's internal reasoning capabilities (ThinkingLevel.MEDIUM) to plan and refine its approach before answering.
Related MCP server: Gemini Research MCP Server
Google Cloud Project: You must have a Google Cloud project with the Vertex AI API enabled.
Authentication: You must have gcloud CLI installed and authenticated:
gcloud auth application-default loginPermissions: Your account needs the Vertex AI User role on the project.
The server requires the following environment variables:
Variable | Description | Default |
| Your Google Cloud Project ID (Required) | - |
| Vertex AI location |
|
Note: GOOGLE_CLOUD_PROJECT can also be used instead of GCP_PROJECT_ID.
npm install
npm run buildAdd the following to your ~/.config/goose/profiles.yaml (or manage via the Goose UI):
gemini-agent:
cmd: node
args:
- /path/to/gemini-agent-mcp/build/index.js
envs:
GCP_PROJECT_ID: "your-project-id"
GCP_LOCATION: "global"Add to your claude_desktop_config.json:
{
"mcpServers": {
"gemini-agent": {
"command": "node",
"args": ["/path/to/gemini-agent-mcp/build/index.js"],
"env": {
"GCP_PROJECT_ID": "your-project-id",
"GCP_LOCATION": "global",
"PATH": "/usr/local/bin:/usr/bin:/bin"
}
}
}
}ask_gemini_agentA single powerful entry point for complex queries.
Arguments: query (string)
Description: Handles research, data analysis, and technical questions by orchestrating search, web page reading, and code execution.
Gemini 3.1 Preview: Uses the gemini-3-flash-preview model; availability may vary by region.
Python-only Code Execution: The code execution environment is restricted to standard Python libraries provided by the Gemini sandbox.
Stdio Transport: This server currently only supports standard I/O communication.
MIT
An AI agent capable of searching the web (search grounding), analyzing specific URLs, and writing/executing Python code to solve complex problems, perform data analysis, or answer research questions.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | The task, question, or research topic for the agent to resolve |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description openly lists the agent's capabilities (web search, URL analysis, code execution). Without annotations, it carries the full burden and does so fairly well, though it omits potential limitations or failure modes.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that efficiently conveys the tool's purpose. It is concise but not overly terse; however, it could be broken into shorter sentences for clarity.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the single parameter and no output schema or sibling tools, the description provides sufficient context to understand when and how to use the tool. It covers the agent's main capabilities.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The sole parameter 'query' is well-described in the schema as 'the task, question, or research topic.' The tool description reinforces this by explaining the agent's actions, adding context beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the agent's capabilities: web search, URL analysis, and Python code execution. It specifies the resource (Gemini agent) and actions, making the purpose unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context on when to use the tool ('complex problems, data analysis, research questions'). However, it does not explicitly state when not to use it or list alternatives, though no sibling tools exist.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v2.0.0ask_gemini_agentWith only one tool, there is no risk of confusion between tools; the single tool's purpose is clearly described.
The single tool name follows a clear verb_noun pattern and is consistently styled, so no inconsistency arises.
One tool is borderline for a server that aims to provide web search, URL analysis, and code execution; typically such capabilities yield at least 3-5 tools.
The tool covers a broad range of tasks, but bundling them into one tool may hide missing operations; a more modular surface would be expected for a domain like this.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Hosted MCP server connecting claude.ai, ChatGPT and other AI apps to your own computer
Official MCP server for Agentwork — delegate tasks to AI agents with human-in-the-loop
Driflyte MCP server which lets AI assistants query topic-specific knowledge from web and GitHub.
An MCP server that gives your AI access to the source code and docs of all public github repos
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/cybaea/google-search-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
