Wandesk - Build Your Own AI Desktop

@atulsharmx a civic complaint tool is a solid use of it, and yeah, context persistence is the hard part.

Honest answer on contradictory context: there's no explicit priority arbitration yet. It mostly comes down to the model's own judgment. For ordinary, surface-level contradictions the model can usually catch and reconcile them. For large-scale or deeper ones it still drifts, and there a human needs to step in to spot it and clean up the memory.

So for now it relies a lot on model intelligence plus you keeping the memory tidy. Explicit priority/scoping rules are something we want to add. Good edge case — thanks.

@praveen_sanapala fair stance — being straight with you: Wandesk does use a shell to operate locally, that's how it builds and runs apps, so it's not sandboxed away from your machine.

What bounds it:

- 100% local — nothing goes to us; it runs through your own model API key.

- Every action is visible — you see each command it runs in the chat, it's not a black box.

- It asks before destructive operations (delete / overwrite / etc.).

So the control is transparency + you in the loop + nothing leaving your machine — not a jail. If "AI touching my system at all" is a hard no, that's a totally fair call.

@kalibdar_guide right now apps run locally inside your Wandesk workspace — they're not web-deployable like V0/Vercel, and there's no built-in "publish & share" flow yet.

An app is just a folder of files (UI + backend + an APP.md), so you can hand those files to someone, but a smooth share/import path isn't there yet. It's a natural next step though.

Different model from V0 really — local-first vs web deploy. Glad you want it on here 🙏

@angga_jiyan_fajar_imanuddin thanks! Honest take: we provide a baseline structure and guidance that keeps each app cleanly separated (UI / logic / data, plus an APP.md the AI reads before editing). But it's not enforced — the whole thing is AI-editable, so how clean it stays really depends on how complex the app gets.

For small, focused apps it holds up well. For genuinely complex ones, our current approach is fairly basic and we're still figuring out better architecture and patterns there.