mentor-mcp-server

Overview

The mentor-mcp-server is a comprehensive implementation of the Model Context Protocol that leverages Deepseek's reasoning capabilities to provide second opinions for LLM agents. It offers four main categories of tools: code analysis (reviews, bug detection, performance optimization), design critique (UI/UX, architecture, accessibility), content enhancement (writing feedback, documentation review), and strategic planning (feature brainstorming, feasibility analysis). The server is well-structured with TypeScript implementation and comprehensive documentation.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

• code_review

  Review and analyze code for bugs, style, and best practices

  Try:Review this TypeScript code for potential security issuesTry:Analyze the performance of this Python implementation
• design_critique

  Provide feedback on UI/UX design and architecture

  Try:Critique this dashboard design for user experienceTry:Analyze the accessibility of this mobile app design
• writing_feedback

  Analyze and improve written documentation and content

  Try:Give feedback on this API documentationTry:Improve the clarity of this user guide
• brainstorm_enhancements

  Generate feature ideas and improvement suggestions

  Try:Brainstorm ways to improve the checkout flowTry:Suggest features for a task management app

Comparable tools

Installation

# Clone the repository
git clone git@github.com/cyanheads/mentor-mcp-server.git
cd mentor-mcp-server

# Install dependencies
npm install

# Build the project
npm run build

{
  "mcpServers": {
    "mentor": {
      "command": "node",
      "args": ["build/index.js"],
      "env": {
        "DEEPSEEK_API_KEY": "your_api_key",
        "DEEPSEEK_MODEL": "deepseek-reasoner",
        "DEEPSEEK_MAX_TOKENS": "8192",
        "DEEPSEEK_MAX_RETRIES": "3",
        "DEEPSEEK_TIMEOUT": "30000"
      }
    }
  }
}

FAQ

What Deepseek model does this server use?

By default, it uses 'deepseek-reasoner' but can be configured to use other Deepseek models via the DEEPSEEK_MODEL environment variable.

Can I use this MCP server with any MCP-compatible client?

Yes, it's compatible with any MCP client that supports tool calling, including Claude Desktop and IDEs with MCP integration.

Compare mentor-mcp-server with