jrklaus8 tightens who can see what - and offers an in-house AI option

Two changes in the Canada fork: every database request now runs under the individual user's permissions, and the AI engine can run entirely on a local machine.

securitymulti-tenant

jrklaus8 reworked the fork so that data access is scoped to each logged-in user rather than running through a single all-powerful account that could, in principle, read everyone's records. In plain terms: the system now enforces its own walls between clients on every request, closing a gap where one tenant's data could leak into another's view. That passthrough is wired through every part of the app that touches the database.

The second move adds the option to run the AI model locally instead of calling an outside service - useful for firms that don't want client material leaving their own hardware. It's a clean, self-contained add-on rather than a rip-and-replace.

So what Anyone evaluating Mike forks for a multi-client or privacy-sensitive practice should look here - this is the kind of isolation and on-premise control that compliance teams ask about first.

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 jrklaus8/mike-Canada, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
86f1ef02 refactor: Add Ollama local inference, RLS middleware, and extract chatTools schemas MikeOSS Bot 2026-05-22 ↗ GitHub
0a309b61 security: Enforce RLS JWT passthrough across all routes. refactor: extract parsers. MikeOSS Bot 2026-05-22 ↗ 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-548.md from inside the repo you want the changes in.

⬇ Download capture-thread-548.md