Last updated: July 20, 2026
Nyquest compresses LLM prompts entirely within your browser or on your own infrastructure. We collect no user data.
This policy covers both the Nyquest Chrome Extension and the Nyquest Self-Hosted Proxy Engine.
Nyquest does not collect, transmit, store, or share any user data. Specifically:
The Nyquest Chrome Extension intercepts outbound API requests to supported LLM services (Claude, ChatGPT, Gemini, Grok) and compresses the text content using 281 pattern-matching rules that run entirely in your browser's JavaScript engine.
All processing happens locally. Your prompts never leave your browser except to the LLM provider you are already using. The compressed text is semantically equivalent to the original — only filler words, verbose phrases, and redundant formatting are removed.
The self-hosted proxy engine operates the same way: all compression happens on your own machine. No data is routed through Nyquest servers.
| Permission | Why |
|---|---|
| storage | Save your compression level preference locally |
| activeTab | Inject compression script into supported LLM chat pages |
| Host access | Intercept and compress outbound API requests on claude.ai, chatgpt.com, gemini.google.com, and grok.com |
The Chrome extension runs a content script in the page's MAIN world (rather than the default ISOLATED world) because it needs to intercept fetch(), XMLHttpRequest, and WebSocket calls made by the LLM chat interfaces.
This is the only way to compress prompt text before it is sent to the provider's servers. The script does not read responses, access cookies, or modify page behavior beyond compressing outbound prompt text.
Nyquest does not integrate with any third-party analytics, advertising, or data collection services. The extension does not load any external resources.
When you use the Nyquest platform (the web app and API), we store your conversations and related content so the service can function — chat history, memory, artifacts, and billing records. Access to that content is restricted to a small number of authorized personnel, who may access it only when necessary to:
Every such access to conversation content is recorded in an internal audit log (who accessed what, and when). We do not sell your content, and we do not use it to train third-party models. If you use your own provider keys (BYOK), your prompts are sent directly to that provider under their terms; we still apply the same access restrictions to any content stored on our systems.
To understand how the app is used and where it fails, the Nyquest web app records a small set of first-party product events — for example that the app was opened, a feature or theme was selected, a message was sent, or a page crashed. These events contain metadata only: they never include the text of your messages, your files, or anything you type. They are stored on our own infrastructure, are not shared with or sent to any third party, and are used solely to improve the product and diagnose problems.
Anonymous visitors are identified only by a random browser-generated identifier, not by name or account.
The nyquest.ai website logs standard web server access data (IP address, user agent, pages visited) for security and operational purposes. This data is stored on our own infrastructure and is not shared with any third parties.
If you submit a demo request, your name, email, and company are stored to process your request. This information is not sold or shared.
If this policy changes, the extension version will be updated and the new policy will be published at this URL. Material changes will be noted in the Chrome Web Store listing.
For privacy questions or concerns: