StrawHat AI Development Repository
@StrawHatAI
I am doing rapid R&D using AI, AI Agents, Claude MCP Servers and VS Code. My findings will go here. I hope you get inspired and build epic features with the tools in this Github.
Overview
What is StrawHat AI Development Repository?
A collection of instructions and links to set up a local AI development environment using tools like Ollama, Open-WebUI, Claude Desktop with MCP Servers, Docker, and VS Code extensions. It is for developers who want to run LLMs locally and build coding AI agents.
How to use StrawHat AI Development Repository?
Install Ubuntu via container, WSL, or USB. Then install Ollama for local LLMs, Open-WebUI for a multi‑LLM frontend, and download models from Ollama’s library. Afterwards, download Claude Desktop and install claude-dev-tools using instructions in the repo. Optionally set up Docker Desktop, Cline AI Agent, or Microsoft’s AI Toolkit for VS Code.
Key features of StrawHat AI Development Repository
- Guides for Ollama local LLM backend
- Open-WebUI multi‑LLM frontend setup
- Claude Desktop + MCP Servers integration
- Docker Desktop for easy hosting environment
- Cline AI Agent as VS Code extension
- AI Toolkit by Microsoft for VS Code
Use cases of StrawHat AI Development Repository
- Running LLMs locally for privacy and offline use
- Building a coding AI agent with Claude Desktop and MCP
- Using Google AI Studio to analyze and summarize videos
- Hosting AI applications in containers via Docker
- Connecting multiple LLMs through Open-WebUI
FAQ from StrawHat AI Development Repository
What is needed to start?
Ubuntu (container, WSL, or USB install), Ollama, Open-WebUI, and Claude Desktop with MCP servers.
How do I verify that the setup works?
You should see a live interface at http://localhost:8080/c/(API_KEY) after following the instructions.
Which parts are tested and which are untested?
The core workflow (Ollama, Open-WebUI, Claude Desktop + MCP) is marked "Tested and working". LiteLLM and static IP hosting are listed as "Untested".