Manuscript Reader, Prose Analysis, and the Query Letter builder are all designed from the ground up to be private by default. Everything happens inside your browser — no text, no manuscripts, and no query drafts are ever uploaded, stored, or shared.
We collect anonymous page-visit counts using GoatCounter, an open-source, privacy-friendly analytics tool. GoatCounter does not use cookies, does not track individuals across sessions, and does not collect personal data. It records only that a page was visited — not who visited it, not what they typed, and not what they listened to.
GoatCounter's full privacy policy is available at goatcounter.com/help/privacy.
Manuscript Reader (read-aloud), Prose Analysis (manuscript linting), and the Query Letter builder all operate entirely client-side. No text leaves your device at any point.
Manuscript Reader uses the Kokoro text-to-speech model (Apache 2.0 licence), which runs entirely inside your browser via WebAssembly and WebGPU. The model is downloaded once from a public CDN and then cached locally on your device by the service worker. No text is ever sent to an external server to generate audio.
The only third-party service this site contacts is GoatCounter for anonymous page-visit counting. No advertising networks, no social media trackers, no session-recording tools.
Because we collect no personal data, there is nothing to retain or delete. The anonymous visit counts held by GoatCounter contain no information that could identify you.
If we ever change our data practices in a meaningful way, we will update the date at the top of this page. Given the architecture of this tool, any meaningful change would require fundamental changes to how it works.
Questions? Use the feedback form on the main page.