MCP Server Logical Solver

🚀 MCP Server Logic Solver

The MCP Server Logic Solver is a powerful tool designed to handle First-Order Logic problems. By integrating natural language understanding and formal logical reasoning, it can verify given premises and conclusions and return results as "True", "False", or "Uncertain".

🚀 Quick Start

Prerequisites

1. Install the MCP-Logic Server
   • Download and compile the source code.
   • Configure the runtime environment variables.
2. Configure the API Interface
   • Set the API access key.
   • Conduct connection debugging tests.
3. Prepare Input Data
   • Ensure the format is correct.
   • Perform syntax and semantic checks.

Installation and Configuration

1. Environment Variable Configuration Create a .env file with the following content:

   DEBUG=False
   API_KEY=your_api_key_here
   

2. MCP Configuration File (mcp_config.json) Modify the example file mcp_config_example.json to the actual mcp_config.json:

   {
     "mcpServers": {
       "mcp-logic": {
         "command": "uv",
         "args": [
           "--directory", 
           "/path/to/mcp-logic/src/mcp_logic",
           "run", 
           "mcp_logic", 
           "--prover-path", 
           "/path/to/mcp-logic/ladr/bin"
         ]
       }
     }
   }
   

Running the Project

1. Single Instance Run Execute the following command in the terminal:

   python test.py
   

   The output results will be saved as output.json.
2. Batch Processing Use the following command for batch processing:

   python main.py input.json output.json
   

   Input Format Description:

   [
     {
       "前提": "Natural language description...",
       "前提-FOL": "First-order logic expression...",
       "结论": "Target conclusion...",
       "结论-FOL": "Corresponding first-order logic form..."
     }
   ]
   

✨ Features

First-Order Logic Input Requirements

To ensure the consistency and correctness of the input, the MCP server needs to follow the following format specifications:

1. Premises and Conclusions
   • Natural Language Form: Describe in concise and clear Chinese.
   • First-Order Logic Form: Use standard symbols and terms, such as ∀, ∃, ¬, ∧, ∨, etc.
2. Symbol Conversion Rules
   • ∀ → all
   • ∃ → exists
   • ¬ → not
   • ∧ → and
   • ∨ → or
3. Input Example

   [
     {
       "前提": "All humans are mortal",
       "前提-FOL": "∀x（Human(x) → Mortal(x))",
       "结论": "Socrates is human, so he is mortal",
       "结论-FOL": "Human(Socrates) ∧ Mortal(Socrates)"
     }
   ]
   

Key Components

1. Logical Solver
   • Function: Responsible for parsing and verifying first-order logic expressions.
   • Input: Natural language descriptions and corresponding first-order logic expressions.
   • Output: Boolean results (True/False/Uncertain) and detailed explanations.
2. Templates
   • Function: Define interface specifications and interaction protocols.
   • Content:
     • Input data format
     • Output data structure
     • Error handling mechanism
3. Parser
   • Function: Convert natural language into first-order logic expressions.
   • Supported Languages: Chinese, English.

📦 Installation

Environment Variable Configuration

Create a .env file with the following content:

DEBUG=False
API_KEY=your_api_key_here

MCP Configuration File (mcp_config.json)

Modify the example file mcp_config_example.json to the actual mcp_config.json:

{
  "mcpServers": {
    "mcp-logic": {
      "command": "uv",
      "args": [
        "--directory", 
        "/path/to/mcp-logic/src/mcp_logic",
        "run", 
        "mcp_logic", 
        "--prover-path", 
        "/path/to/mcp-logic/ladr/bin"
      ]
    }
  }
}

💻 Usage Examples

Single Instance Run

python test.py

The output results will be saved as output.json.

Batch Processing

python main.py input.json output.json

Input Format Description:

[
  {
    "前提": "Natural language description...",
    "前提-FOL": "First-order logic expression...",
    "结论": "Target conclusion...",
    "结论-FOL": "Corresponding first-order logic form..."
  }
]

📚 Documentation

Output Format

The generated output file will contain the following fields:

{
    "前提": "Original natural language description",
    "前提-FOL": "Processed first-order logic expression",
    "结论": "Target conclusion",
    "结论-FOL": "Corresponding first-order logic form",
    "答案": "True/False/Uncertain",
    "解释": "Detailed reasoning process",
    "工具调用": "True/False"
}

Troubleshooting

1. MCP Server Connection Issues
   • Confirm that the server is running.
   • Check if the port configuration is correct.
   • Check the network connection status.
2. Input Data Errors
   • Check the format specifications.
   • Perform syntax checks.
   • Ensure content consistency.

Through the above steps, you can efficiently use the MCP Server Logic Solver to solve various complex problems.