Tianji Thinking Models

What is "Tianji" Thinking Models MCP Server?

"Tianji" (天机) Thinking Models MCP Server is a powerful Model Context Protocol (MCP) based service for querying and managing thinking models. Named after the ancient Chinese concept meaning "Heaven's secrets," Tianji provides a comprehensive collection of thinking model data, including model definitions, purposes, categories, popular science teaching content, usage notes, and more. This specialized MCP service enables users to easily access and integrate structured thinking methodologies into AI assistants and knowledge systems.

What are the key features of "Tianji" MCP Server?

• Rich Thinking Model Library: Contains classic thinking models across multiple domains including decision theory, systems thinking, and probabilistic thinking
• Multi-step Reasoning Support: Assists language models in performing multi-step reasoning tasks for complex problems, even for models without strong native reasoning capabilities
• MCP Protocol Integration: Seamlessly integrates with AI assistants through the Model Context Protocol
• Bilingual Support: Complete Chinese and English data support
• Intelligent Model Recommendations: Automatically recommends suitable thinking models based on problem characteristics
• Interactive Reasoning Process: Guides users through structured thinking processes
• Learning System: Continuously improves through user feedback
• Model Creation Tools: Allows creation of new models or combining existing ones

What are the use cases of "Tianji" MCP Server?

1. AI Assistant Enhancement

   • Integrating systematic thinking into AI conversations
   • Providing structured problem-solving frameworks
   • Enhancing AI reasoning capabilities through MCP integration

2. Education and Training

   • Building thinking model knowledge bases
   • Developing cognitive training courses
   • Designing educational tools with "Tianji" integration

3. Knowledge Management Systems

   • Integrating "Tianji" MCP services
   • Establishing organizational knowledge bases
   • Supporting decision analysis systems with thinking models

4. Personal Growth

   • Systematic learning of thinking models
   • Improving cognitive decision-making
   • Developing critical thinking skills through guided processes

FAQ

How do I access the "Tianji" thinking model data?

Data files are stored in the thinking_models_db directory:

• zh/: Chinese thinking model data files (天机中文思维模型)
• en/: English thinking model data files

What MCP tools does "Tianji" provide?

The "Tianji" MCP Server provides four categories of tools:

1. Exploration Tools

   • list-models: List all thinking models
   • search-models: Search models by keywords
   • get-model-info: Get detailed model information
   • get-categories: Get all model categories

2. Problem-Solving Tools

   • recommend-models-for-problem: Get model recommendations
   • interactive-reasoning: Interactive reasoning guidance
   • explain-reasoning-process: Explain model reasoning process

3. Creation Tools

   • create-thinking-model: Create new thinking models
   • emergent-model-design: Combine existing models

4. System Tools

   • get-server-version: Version information
   • count-models: Statistical information

How do I switch languages in "Tianji"?

"Tianji" is fully bilingual. Specify the language code via the lang parameter when calling any MCP tool:

• zh: Chinese (default) - 中文
• en: English

How do I integrate "Tianji" MCP Server?

"Tianji" offers multiple integration methods:

1. NPX Method (Recommended)

   {
     "mcpServers": {
       "thinking-models": {
         "command": "npx",
         "args": ["--no-cache", "@thinking-models/mcp-server"]
       }
     }
   }
   

2. Local Node.js Method

   {
     "mcpServers": {
       "tianji": {
         "command": "node",
         "args": ["path/to/thinking_models_server.js"]
       }
     }
   }
   

How do I start developing with "Tianji"?

1. Clone the repository: git clone https://github.com/lanyijianke/thinking_models_mcp.git
2. Install dependencies: npm install
3. Build the project: npm run build
4. Start the server: node build/thinking_models_server.js

License

MIT License