Token Optimizer

Token Optimizer is a focused side-panel extension that helps you turn rough, long, or repetitive prompts into cleaner input for Gemini™ or ChatGPT before you send anything. Instead of pasting a messy prompt directly into an assistant, open Token Optimizer beside Gemini™ or ChatGPT, capture or paste your prompt, and choose Prepare only or Prepare & insert. Preparation does not call another AI model. The extension does not auto-send messages; you stay in control and review the final prompt first. This is designed for people who use AI tools for coding, research, writing, planning, debugging, and multi-step work. It helps reduce prompt clutter, remove repeated instructions, preserve the actual task, and keep the final prompt easier for the assistant to follow. Key features: - Capture prompt text from Gemini™ or ChatGPT, or paste it manually. - Optimize long, repetitive, or messy prompts into concise assistant-ready input. - Prepare prompts without making a duplicate provider model call. - Insert the optimized prompt into the active assistant only after you choose to do so. - Keep the workflow visible with simple stages: Capture, Prepare, Ready, Insert, Review. - Avoid auto-sending messages so you can inspect the final prompt before submitting. - Keep the extension narrow and focused on prompt cleanup for supported assistants. How it works: 1. Open Gemini™ or ChatGPT in Chrome. 2. Open the Token Optimizer side panel. 3. Paste your rough prompt or capture text from the assistant prompt box. 4. Click Prepare only, or use Prepare & insert for the one-click path. 5. Review the cleaned assistant-ready prompt. 6. Copy it or insert it into the active assistant. 7. Send only when you are ready. Privacy and data use: - A prominent in-product notice explains prompt, local-metric, and operational request-data handling before any prompt is typed, pasted, captured, or prepared. - Prompt input and capture stay disabled until you check the consent box. Prompt text is sent over HTTPS to Token Optimizer only after that consent and a Prepare action. - The preparation endpoint uses deterministic processing and does not call a provider model. - Raw and prepared prompt text is processed for the response and is not stored by Token Optimizer. - Nothing is sent when you capture text or insert text into an assistant. - The extension does not auto-send assistant messages. - The extension does not store provider API keys. - The extension does not sell user data. - Token counts and strategy are shown only in the current side-panel session; no prompt text or usage history is stored. Token Optimizer is not affiliated with, endorsed by, or sponsored by Google or OpenAI. Gemini is a trademark of Google LLC. Use of this trademark is subject to Google Permissions.