FAQ

Tokavy is a desktop utility that refines your AI prompts in one keystroke. You highlight any text in any app — whether it's Claude, ChatGPT, Cursor, or Notion — and press a shortcut. Tokavy either Compresses your prompt (removes filler words, makes it tight and direct) or Enhances it (adds context, structure, and role-framing so the AI responds with more precision). The refined prompt replaces your original instantly.

Tokavy works in any application where you can type or select text. This includes Claude, ChatGPT, Gemini, Cursor, Windsurf, Bolt, Lovable, Replit, Notion, VS Code, GitHub Copilot, Perplexity, Poe, and any browser-based or desktop tool. If you can highlight text in it, Tokavy works in it — no integrations required.

Compress (Ctrl + .) strips your prompt down to its essential intent. It removes filler words, hedging language, and redundancy. Use it when your prompt is too wordy and you want a faster, cheaper API call.

Enhance (Ctrl + M) expands and restructures your prompt. It adds role context, output format instructions, and logical flow. Use it when you need the AI to produce a more detailed, accurate, or structured response.

It depends on how verbose your original prompt is, but most users see 30–55% token reduction with Compress mode. A 200-token messy prompt typically compresses to around 80–120 tokens with no loss of intent. For heavy API users running hundreds of calls a day, this directly reduces your OpenAI / Anthropic / Gemini bill.

Your prompt is sent to Tokavy's backend API to process the refinement — it is not stored, logged, or used for training. We process and discard. We will never sell or share your prompt data. You can verify this in our Privacy Policy. If you require full on-device processing with zero network calls, that's on our Pro roadmap.

Tokavy is currently available on Windows 10 and 11. macOS and Linux builds are in development and will be released as part of an upcoming update. Sign up for the Pro waitlist to get notified when your platform is supported.

The Free plan gives you 25 refinements per day — enough for casual use. The Pro plan gives you 200 refinements per day, priority model routing (faster responses, better refinement quality), and early access to new features as we ship them. Both plans include Compress, Enhance, and toast notifications.

Yes, absolutely. There are no contracts and no cancellation fees. You can cancel anytime from your account dashboard, and you'll retain Pro access until the end of your current billing period. After that, you drop back to the Free plan — your account and data stay intact.