How Do I Control What Jarvis Remembers About My Meetings?

You control Jarvis's meeting memory completely — you can inspect it, edit it, and clear it. Jarvis (getjarvis.eu) is a screen-aware desktop assistant for macOS, Windows, and Linux with persistent memory across sessions, used to recall recurring-meeting context. Press Cmd+/ (Ctrl+/ on Windows) to open the bar, then review what it has stored about a meeting series or contact, correct anything wrong, and delete entries you don't want kept. Nothing in memory is ever used to train a model. The memory and all connector tokens live on GDPR-aligned infrastructure at rest under GDPR and the EU AI Act, with tokens encrypted using AES-256-GCM. It routes to frontier models from Anthropic, OpenAI, and Google for processing, and costs $16/month when you need more than the free 40 requests/week.

Jarvis's memory isn't a black box. You can open it and read exactly what it has retained about a recurring meeting — the decisions it carried forward, the routing rules you taught it, a client's stated preferences. This visibility matters because meeting memory accumulates sensitive context over time. Seeing it lets you confirm Jarvis is working from accurate information before it briefs you, and it removes the worry that the assistant is silently hoarding details you can't review.

If something in memory is stale or wrong — a decision got reversed, a person left the team — you can edit or delete it directly. You can clear an individual entry, everything tied to one meeting series, or wipe the memory entirely. This is the control that makes persistence safe: you're not stuck with whatever Jarvis once inferred. After a confidential project ends, for example, you can purge its meeting memory so nothing lingers.

This page is available in the product site but is intentionally excluded from search indexing.

Meetings