Support & Help

Set up vloop

• Desktop app — install the local Desktop Hub, then connect from inside the app (preview → confirm → write). It is an unsigned local build for now.
• CLI — run vloop in a project; vloop doctor checks your setup; vloop connector add sets up the Claude Desktop connector.
• Claude Desktop — add the vloop connector; Claude calls the framing tools and answers. vloop frames; it never answers.
• Browser AI bridge — load the local vloop browser extension (unpacked) and enable it on a supported site; vloop frames in the page and you send it.
• ChatGPT — the official connector path is planned, not live. Once a supported path exists you link it and ChatGPT answers.
• Mobile hub — open the vloop mobile app (a connector hub) and tap a model to open it; the model answers in its own app.

Known limitations (private alpha)

• This is a private/test build. Public release, app-store listings, and signed installers are not available yet.
• The real-model demo is limited to allowlisted alpha accounts; everyone else uses the mock path.
• Free authenticated demo use is capped with a small lifetime limit to prevent abuse.
• Account, billing, and history are not enabled yet.
• The ChatGPT connector and a hosted endpoint are planned, not live.

Privacy summary

• vloop keeps no prompts, answers, or history — comparisons are request–response only.
• When you choose a live provider, your prompt is sent server-side to that provider so the answer can be generated.
• Provider API keys stay server-side and are never sent to your browser.
• No telemetry and no long-term user model — vloop does not profile you.
• The account layer stores only identity and a content-free usage counter — never prompts or answers.

This is a plain-language summary of what is true today, not a final legal privacy policy. A formal privacy policy and terms will accompany any public release.

Contact

• A support contact is not finalized for this private alpha. Please reach out through the channel you were invited from.