manueljpconde lets firms keep Mike's brain in-house

The mikeEU fork can now point the assistant at a self-hosted AI model instead of a cloud provider.

infrastructuresecurity

Most legal-AI tools send your prompts - and whatever documents ride along with them - out to a third-party model in someone else's data centre. manueljpconde adds a path that keeps it local: an operator can wire Mike up to an AI model running on the firm's own servers, configured once per deployment and shared by everyone on that instance. Per-user setups are flagged as coming later.

There's a nice bit of honesty engineering, too. Smaller self-hosted models often can't run Mike's document and drafting tools. Rather than let the AI quietly claim it did something it couldn't, the team hard-codes a warning into the model's instructions so it won't invent actions it never took - a real guard against the confident-but-fake answers that erode trust in these systems.

So what Worth a look for any firm or legal-ops team whose confidentiality or data-residency rules make sending matter data to a public AI provider a non-starter.

View this fork on GitHub →

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

Commits in this thread

2 commits from manueljpconde/mikeEU, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
c0361944 feat: add self-hosted local llm provider Manuel Conde 2026-05-09 ↗ GitHub
c8ef360f Merge pull request #2 from manueljpconde/codex/local-llm-provider Manuel Conde 2026-05-09 ↗ GitHub
Add self-hosted local LLM provider support

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-218.md from inside the repo you want the changes in.

⬇ Download capture-thread-218.md