​
Run Mode

scmcphub provides two distinct run modes to accommodate different user needs and preferences.

​
Tool Mode

In tool mode, scmcphub provides a curated set of predefined functions that the LLM can select and execute.

Advantages:

  • Stable: Predefined functions ensure consistent and reliable execution
  • Predictable: Known behavior and expected outputs
  • Safe: Controlled environment with validated operations

Disadvantages:

  • Limited flexibility: Restricted to available predefined functions， you should define new tools when you need customization functions

​
Usage

Running in terminal

scmcp run --run-mode tool

Below are some examples:

Configure mcp

{
  "mcpServers": {
    "scmcp": {
      "command": "/home/test/bin/scmcp",
      "args": [
         "run", "--run-mode", "tool"
      ]
    }
  }
}

​
Code Mode

In code mode, scmcphub provides a Jupyter backend that allows the LLM to generate and execute custom code.

The mode is based on the project https://github.com/huang-sh/abcoder

Advantages:

  • Highly flexible: Can create custom workflows and combine operations freely
  • Extensible: Supports any Python code and external libraries
  • Interactive: Real-time code execution and debugging capabilities

Disadvantages:

  • Less stable: Code generation is different for each time

​
Usage

Running in terminal

scmcp run --run-mode code

Configure mcp

{
  "mcpServers": {
    "scmcp": {
      "command": "/home/test/bin/scmcp",
      "args": [
         "run", "--run-mode", "code"
      ]
    }
  }
}

Below are some examples: https://youtu.be/3jtXIeapslI