manueljpconde lets you bring your own AI model - and locks the keys away
This fork opens the door to private and self-hosted AI engines, so a firm's data and credentials never have to leave the building.
Until now, the app leaned on a fixed set of cloud AI providers. manueljpconde adds a "managed models" option: any team can point the app at its own AI endpoint - a private corporate model, Microsoft's Azure-hosted AI, or even a model running on the user's own machine - and supply its own key. Those keys are encrypted before they're stored and are never handed back to the browser, which matters for anyone nervous about credentials leaking.
The other half is setup. The fork ships a one-command local install that brings up the whole system - app, database, file storage, sign-in - with placeholder settings, so a curious firm can spin up a working private instance to kick the tyres without wiring up any cloud accounts first. The accompanying notes are careful to spell out which keys go where, a frequent stumbling block for Azure shops.
Spotted something wrong? Or know the PR text has fresher detail than the writeup above?