andrsschultz moves Mike's AI engine in-house

This fork cuts the cords to outside AI services and runs the model on hardware you control.

infrastructuresecurity

andrsschultz tore out Mike's connections to the big cloud AI providers - Anthropic's Claude and Google's Gemini - along with the screens where each user pastes in their own provider keys. In their place sits a single client pointed at a model running locally, by default the kind you'd host on your own machine with a tool like Ollama (a popular way to run AI models on local hardware).

The plumbing underneath is well done: it handles streaming replies, step-by-step reasoning from capable models, and multi-round tool use without leaning on any outside service. The trade-off is that this version assumes a local-only setup, so it's built for self-hosting rather than a shared, hosted deployment.

So what Worth a look for any firm or legal team that can't send client material to third-party AI services and wants the model running entirely on premises.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

1 commit from andrsschultz/mike, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
5b79b283 Route LLM calls to local provider asschultz 2026-04-30 ↗ GitHub

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-82.md from inside the repo you want the changes in.

⬇ Download capture-thread-82.md