Gemini MCP Server
# Gemini MCP Server
Model Context Protocol (MCP) server implementation that enables Claude Desktop to interact with Google's Gemini 2.5 Pro Experimental AI model.
## Features
- Full MCP protocol support
- Google Gemini 2.5 Pro Experimental model access
- Secure API key handling
- Google Search integration (optional)
- Token usage reporting
- TypeScript implementation
## Available Tools
The MCP server provides the following tools:
1. **generateWithGemini** - Generate content with Google Gemini 2.5 Pro Experimental
- Parameters:
- `prompt` (string, required): The prompt to send to Gemini
- `temperature` (number, optional): Temperature setting (0.0 to 1.0)
- `maxTokens` (number, optional): Maximum output tokens
- `safeMode` (boolean, optional): Enable safe mode for sensitive topics
- `useSearch` (boolean, optional): Enable Google Search grounding tool
2. **getModelInfo** - Get information about the Gemini model being used
## Common Issues
1. **Connection Issues**
- Check that you have a valid API key
## Security
- API keys are handled via environment variables only
- No sensitive data is logged or stored
TDQS
Scored across 2 tools
The two tools have completely distinct purposes: generateWithGemini is for content generation, while getModelInfo is for retrieving model metadata. There is no overlap in functionality, making it impossible to confuse them.
Both tools use camelCase naming, which is consistent. However, generateWithGemini uses a verb+preposition+noun pattern, while getModelInfo uses verb+noun, creating a minor deviation in structure.
With only 2 tools, the server feels thin for a Gemini API server, as it lacks operations like listing models, managing conversations, or handling multimodal inputs. The count is borderline for the apparent scope of interacting with a generative AI model.
The tool surface is severely incomplete for a Gemini API server. It only covers content generation and model info, missing essential operations such as chat/completion management, file uploads for multimodal inputs, or configuration settings, which will likely cause agent failures in complex workflows.