Happy slice of mcp pizzaMCP.pizza

OllamaAssist

MCP.Pizza Chef: madtank

OllamaAssist is a powerful MCP client designed to run AI assistants locally using Ollama's LLMs, optimized for Llama models but adaptable to any function-calling capable language model. It features full MCP integration, enabling seamless interaction with a universal tool protocol and dynamic tool support through automatic capability detection. The client provides a real-time streaming chat interface built with Streamlit, enhancing user interactivity and responsiveness. OllamaAssist supports advanced function calling and self-reflection, making it ideal for developers seeking to build sophisticated, locally hosted AI assistants with standardized, discoverable tool communication.

Unmaintained · No commits in 20 months.

Use This MCP client To

Run local LLMs with advanced function calling Build interactive AI chat assistants with Streamlit Integrate diverse AI tools via MCP protocol Enable real-time streaming chat interfaces Automatically detect and use tool capabilities Develop self-reflective AI assistant workflows

README

OllamaAssist

A Streamlit interface for Ollama models with full MCP (Model Context Protocol) integration. Works with any tool-calling capable model like deepseek-r1-tool-calling:14b or llama2:latest.

Key Features

  • Local LLM Execution: Run models locally using Ollama (deepseek-r1)
  • MCP Integration: Universal tool protocol support
  • Streamlit Interface: Real-time streaming chat interface
  • Dynamic Tool Support: Automatic capability detection

What is MCP (Model Context Protocol)?

MCP is a universal protocol that standardizes how AI models interact with tools and services. It provides:

  • Universal Tool Interface: Common protocol for all AI tools
  • Standardized Messages: Consistent communication format
  • Discoverable Capabilities: Self-describing tools and services
  • Language Agnostic: Works with any programming language
  • Growing Ecosystem: Many tools available

Learn more:

Prerequisites

  • Python 3.9+
  • Ollama desktop app installed and running
  • MCP-compatible tools
  • python-dotenv
  • An Ollama-compatible model with tool-calling support

Installation

  1. Prerequisites:

    # Install Ollama desktop app from https://ollama.ai/download

# Make sure Ollama is running
# Then pull the recommended model (or choose another tool-calling capable model)
ollama pull MFDoom/deepseek-r1-tool-calling:14b

# Alternative models that support tool calling:
# ollama pull llama2:latest

  2. Setup:

    git clone https://github.com/madtank/OllamaAssist.git
cd OllamaAssist
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

OllamaAssist FAQ

How does OllamaAssist handle local model execution?
OllamaAssist runs language models locally using the Ollama platform, allowing offline and private AI assistant deployment.
Can OllamaAssist work with models other than Llama?
Yes, while optimized for Llama models, OllamaAssist supports any function-calling capable LLM, including models like deepseek-r1-tool-calling:14b.
What interface does OllamaAssist provide for user interaction?
It offers a real-time streaming chat interface built with Streamlit for interactive and responsive AI conversations.
How does OllamaAssist support tool integration?
It fully integrates the Model Context Protocol (MCP), enabling universal, standardized communication with a wide range of AI tools and services.
Does OllamaAssist support dynamic tool capability detection?
Yes, it automatically detects available tool capabilities to optimize AI assistant functionality.
Is OllamaAssist language-specific?
No, MCP and OllamaAssist are language-agnostic, supporting tools and models across different programming languages.
Can OllamaAssist be used with cloud-based LLMs?
While designed for local execution, OllamaAssist can adapt to any function-calling LLM, including cloud-hosted models that support MCP.
How does OllamaAssist enhance AI assistant interactivity?
Through its Streamlit interface and real-time streaming, it provides a smooth, engaging user experience with immediate feedback.