Deep Thinking Assistant - Gemini

マルチエージェント型MCP

html

0

GitHub Stars

html

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "shark-bot-0118-dive-deep-mcp": {
      "command": "python",
      "args": [
        "dive_deep_server.py"
      ],
      "env": {
        "GEMINI_MODEL": "gemini-2.0-flash",
        "GEMINI_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Deep Thinking Assistant is a Gemini API based MCP server that provides deep thinking and analytical capabilities. It integrates with AI editor models to offer enhanced insights, multi-perspective problem analysis, and practical, concrete proposals for improvement.

How to use

You connect to this MCP server using an MCP client to start a session that guides you through deep thinking, analysis, and actionable outputs. Use the deep thinking toolchain to explore problems from multiple angles, request enhancements to code, and obtain a final, critically analyzed review of proposed solutions. Operate with clear instructions, provide context for the analysis, and specify the target model to tailor responses.

How to install

Prerequisites: ensure you have Python 3.8+ and pip installed on your system.

pip --version
python --version

Install dependencies from the requirements file that accompanies the server project.

pip install -r requirements.txt

Configure environment variables for the Gemini API and model used by the server. Create a file named ".env" and set the following values.

GEMINI_API_KEY=your_api_key_here
GEMINI_MODEL=gemini-2.0-flash

Start the MCP server by running the main script.

python dive_deep_server.py

Additional sections

Configuration and runtime details are centered on the following environment variables and runtime script. If you need to customize behavior, adjust the values in your ".env" file and restart the server.

Available tools

deep_thinking_agent

Analyzes problems from multiple angles, identifies key viewpoints, and provides guidelines for deeper understanding and better solutions.

enhancement_agent

Examines code with a focus on quality, performance, and maintainability, offering concrete, practical improvement suggestions.

final_review_agent

Performs a final, critical review of implemented improvements, identifying potential issues and opportunities for further optimization.