This is an implementation of the Model Context Protocol (MCP) that integrates with Google's Gemini API to provide analytical thinking capabilities without code generation.
The Gemini Thinking Server is a specialized MCP server that leverages Google's Gemini model to provide sequential thinking and problem-solving capabilities. It allows for:
- Breaking down complex problems into steps
- Planning and design with room for revision
- Analysis that might need course correction
- Problems where the full scope might not be clear initially
- Gemini-Powered Thinking: Utilizes Gemini's analytical capabilities to generate thoughtful responses
- Meta-Commentary: Provides insights into the reasoning process
- Confidence Levels: Indicates how confident Gemini is in its analysis
- Alternative Paths: Suggests different approaches to the problem
- Branching Thoughts: Allows exploration of different thought paths
- Revision Capability: Supports revising previous thoughts
- Session Persistence: Save and resume analysis sessions
# Clone the repository
git clone <repository-url>
# Install dependencies
npm install
# Build the project
npm run buildBefore running the server, you need to set up your Gemini API key:
export GEMINI_API_KEY=your_api_key_herenode dist/gemini-index.jsThe geminithinking tool accepts the following parameters:
query(required): The question or problem to analyzecontext(optional): Additional context informationapproach(optional): Suggested approach to the problempreviousThoughts(optional): Array of previous thoughts for contextthought(optional): Your current thinking step (if empty, will be generated by Gemini)nextThoughtNeeded(required): Whether another thought step is neededthoughtNumber(required): Current thought numbertotalThoughts(required): Estimated total thoughts neededisRevision(optional): Whether this revises previous thinking
826C
revisesThought(optional): Which thought is being reconsideredbranchFromThought(optional): Branching point thought numberbranchId(optional): Branch identifierneedsMoreThoughts(optional): If more thoughts are needed
The tool also supports session management commands:
sessionCommand: Command to manage sessions ('save', 'load', 'getState')sessionPath: Path to save or load the session file (required for 'save' and 'load' commands)
{
"sessionCommand": "save",
"sessionPath": "/path/to/save/session.json",
"query": "dummy",
"thoughtNumber": 1,
"totalThoughts": 1,
"nextThoughtNeeded": false
}{
"sessionCommand": "load",
"sessionPath": "/path/to/load/session.json",
"query": "dummy",
"thoughtNumber": 1,
"totalThoughts": 1,
"nextThoughtNeeded": false
}{
"sessionCommand": "getState",
"query": "dummy",
"thoughtNumber": 1,
"totalThoughts": 1,
"nextThoughtNeeded": false
}Here's an example of how to use the tool:
{
"query": "How might we design a sustainable urban transportation system?",
"context": "The city has 500,000 residents and currently relies heavily on personal vehicles.",
"approach": "Consider environmental, economic, and social factors.",
"thoughtNumber": 1,
"totalThoughts": 5,
"nextThoughtNeeded": true
}The server responds with:
{
"thought": "The generated thought from Gemini",
"thoughtNumber": 1,
"totalThoughts": 5,
"nextThoughtNeeded": true,
"branches": [],
"thoughtHistoryLength": 1,
"metaComments": "Meta-commentary about the reasoning",
"confidenceLevel": 0.85,
"alternativePaths": ["Alternative approach 1", "Alternative approach 2"]
}Several example clients are provided to demonstrate different use cases:
sample-client.js: Basic client exampleexample-usage.js: Specific usage examplecodebase-analysis-example.js: Example for codebase analysissession-example.js: Example demonstrating session persistenceadvanced-filtering-example.js: Example demonstrating advanced semantic filtering
To run the session example:
node dist/session-example.jsTo run the advanced filtering example:
node dist/advanced-filtering-example.jsMIT