Ollama

Ollama + Python MCP

python

0

GitHub Stars

python

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": {
    "tunamsyar-ollama-mcp-py": {
      "command": "python",
      "args": [
        "simple_modular_server.py"
      ]
    }
  }
}

You run a modular MCP (Model Context Protocol) server implemented in Python that lets you register tools, detect user intent, and execute those tools asynchronously. It is designed for learning and experimentation, serving as a practical way to see how prompts, local language models, and tool invocations can work together in a single server.

How to use

To use this MCP server, start the server process and then interact with it through a client that sends natural language commands. You can ask it to create notes, fetch weather, perform calculations, query the time, run shell commands safely, or perform file operations. The server will detect your intent, map it to the appropriate tool, run the action, and present the results.

Common usage patterns include: creating and searching notes, requesting current weather, asking for computations, querying the current time, executing safe shell commands, and performing simple file operations. You can also combine actions in a single conversation to build complex workflows.

How to install

Prerequisites you need before installing and running the MCP server:

Python 3.8+ must be installed on your system
Ensure Ollama is installed and a supported model (for example phi or llama3.2:3b) is running

Step-by-step setup:

git clone <repository-URL>
cd ollama-mcp-py
python -m pip install -r requirements.txt
python simple_modular_server.py
python client_script.py

Additional notes

This project is intended for learning and experimentation, not production use. It demonstrates how to design modular tools, perform intent detection, and integrate local language models for structured command extraction.

Available tools

note

Create and search notes using an SQLite backend to store and retrieve note content by title or category.

weather

Fetch current weather information for a given location using a weather tool.

calc

Perform mathematical calculations and return results with optional units.

time

Query the current time or compute time-based information for a specified zone.

shell

Safely execute predefined shell commands and return their output.

file_ops

Perform basic file operations such as read, write, and list directory contents.