GemSuite MCP
Comprehensive Gemini API integration for MCP hosts with intelligent model selection and file handling.
Add to your client
Copy the config for your MCP client and paste it into its config file.
npx -y @smithery/cli@latest install @PV-Bhat/gemsuite-mcp --client claudePaste into ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"gemsuite-mcp": {
"command": "npm",
"args": [
"start"
],
"env": {
"GEMINI_API_KEY": "your_api_key_here"
}
}
}
}Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf
Before you start
- Node.js 16+
- A Google Gemini API key from Google AI Studio (set as GEMINI_API_KEY)
About GemSuite MCP
An MCP server that wraps the Google Gemini API and automatically routes each request to the most appropriate Gemini model based on task type, content type, and complexity. It offers four tools (search, reason, process, analyze) plus unified multimodal file handling, aiming to minimize token cost while maximizing capability for MCP hosts like Claude Desktop and Cursor.
Tools & capabilities (4)
gem_searchInformation retrieval with Gemini search integration (Gemini Flash). For factual questions, current information, and grounded responses; supports file inputs and optional thinking.
gem_reasonComplex reasoning with step-by-step analysis (Gemini Flash Thinking). For math, science, coding problems, and logical analysis; supports a file_path and show_steps option.
gem_processFast, token-efficient content processing (Gemini Flash-Lite). Supports operations: summarize, extract, restructure, simplify, expand, critique, feedback, and analyze.
gem_analyzeIntelligent file analysis with automatic model selection. For document analysis, code review, and image understanding via a file_path and instruction.
When to use it
- Summarize or extract data from large documents (PDF, DOCX) efficiently
- Analyze images and code files with auto-selected Gemini models
- Solve complex math, science, or coding problems with step-by-step reasoning
- Answer factual or current-events questions with Google Search grounding
Security notes
Requires a Google Gemini API key (GEMINI_API_KEY), obtainable from Google AI Studio. Store it in a .env file or as an environment variable; do not commit it. File-handling tools accept local file paths, so the server has access to files you reference.
GemSuite MCP FAQ
How do I install GemSuite MCP?
Recommended: run `npx -y @smithery/cli@latest install @PV-Bhat/gemsuite-mcp --client claude`. Alternatively, clone the repo, run `npm install`, set GEMINI_API_KEY in a .env file, run `npm run build`, then `npm start`.
Which Gemini models does it use?
It uses Gemini 2.0 Flash (search, multimodal), Gemini 2.0 Flash-Lite (fast, cost-efficient processing), and Gemini 2.0 Flash Thinking (reasoning), selecting automatically per task. You can override with the model_id parameter.
Do I need an API key?
Yes. Obtain a Gemini API key from Google AI Studio and set it as the GEMINI_API_KEY environment variable or in a .env file.
Which MCP hosts are supported?
Any MCP-compatible host, including Claude Desktop, Cursor IDE, and Replit.
Alternatives to GemSuite MCP
Compare all alternatives →
Official MCP server providing persistent, file-backed knowledge-graph memory across sessions.
Structured step-by-step reasoning tool for breaking problems into revisable thought sequences.
Fully managed remote server for AWS docs, blogs, What's-New and Well-Architected guidance — no key.
Compare GemSuite MCP with: