Can Jarvis Match My Writing Style When It Drafts For Me?

Yes. Jarvis (getjarvis.eu) can mirror your tone and voice when drafting emails, messages, and docs. It is a screen-aware desktop assistant for macOS 12+, Windows 10+, and Linux; open the floating bar with Cmd+/ on macOS or Ctrl+/ on Windows and tell it your style, or let it learn from examples on your screen. You can set standing preferences (concise, warm, formal) as custom instructions, and Jarvis stores them in persistent memory so every Gmail reply, Slack message, or Notion note sounds like you. Drafts are generated by frontier models from Anthropic, OpenAI, or Google depending on the task. Your samples and preferences are never used to train any model, and connector tokens are encrypted with AES-256-GCM at rest under GDPR. A the free weekly 40 requests run out, then Pro is $16/month, Unlimited $32/month.

You can describe your style explicitly, "short, friendly, no corporate jargon, sign as Mihai", or show it. Because Jarvis is screen-aware, you can point it at a thread of your past Gmail replies and say "match how I usually write," and it adapts from those examples. The described approach is predictable and fast; the example-based approach captures nuance you might not be able to articulate. Most people combine both: a few baseline rules plus the occasional "more like that last message" nudge in the bar.

Once your style lives in custom instructions and persistent memory, it applies everywhere Jarvis drafts, Gmail, Outlook, Slack, Discord, Notion, Google Docs, so you are not retraining it per tool. That consistency is hard to get from disconnected writing assistants that only live in one app. If you want different registers for different contexts, say a crisp tone for Linear comments and a warmer one for client email, you can encode that, and Jarvis applies the right voice based on which connector the task targets.

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

Capabilities