Privacy policy

Last updated 8 November 2026

DoomScroll.build turns your reviews (such as GitHub pull requests) into short narrated videos. This page says what we collect, why, and who else sees it. We do not sell your data and we do not track you across other apps or websites.

What we collect

Account details from the provider you sign in with: your GitHub account ID, username, display name and avatar, or your Google or Microsoft identity (name and email address).
Sign-in tokens for the providers you connect, stored on our server so DoomScroll.build can read your reviews and act when you ask it to (for example post a comment or merge). Tokens are never shown in the app.
Review content you can access: pull request titles, descriptions, diffs, comments, and content from other sources you connect. We read it to show your feed and to make videos.
What you do in DoomScroll.build: comments, approvals, video requests, feedback on videos, and which cards you watched and for how long, so your feed knows what is done.
Coding agent integrations: the version of the DoomScroll.build plugin or client you use, so we can tell you when it needs an update. While you are signed in, the Claude Code plugin also sends the session's status (working, waiting for you, turn ended), the names of tools it ran and whether each succeeded, and a short completion summary with links that the agent marks for you. It never sends your prompts, the conversation transcript, tool inputs or outputs, or file paths. Run report off to stop it. Questions the agent asks you are sent only when you turn that on.
Chrome extension: when you click its button or press its shortcut on a GitHub pull request, it sends that pull request's address to DoomScroll.build to request a video. It keeps your DoomScroll.build sign-in key in the browser's extension storage on your device. It does not read other pages, your browsing history or GitHub credentials.
Purchases made on our website go through our payment provider, Polar, which collects your payment details. We keep only what you bought and your credit balance. The iOS app does not sell anything.

Who else processes it

AI providers (Anthropic and OpenAI) receive the review content of a video you or your settings request, to write and narrate that video.
GitHub, Google and Microsoft receive the actions you ask DoomScroll.build to take on your behalf, such as a comment or a merge. A published video is linked from the pull request on GitHub.
Product analytics and session replay (self-hosted PostHog): product events under a pseudonymous ID, plus recordings of how pages are used (clicks, scrolling and what is on screen, which can include review content you view). Text you type is masked, and secrets shown once (like a new API key) are never recorded. Tokens are never recorded. Browsers that send Global Privacy Control or Do Not Track are excluded, and you can switch this off in Settings.
Polar processes website payments as merchant of record.

How long we keep it

We keep your data while your account exists. Deleting your account (Settings, then Delete account) removes your sign-in tokens, sessions, keys and connected sources at once. Records of videos and purchases are kept for billing and abuse prevention but are no longer linked to a usable account. Videos already posted on GitHub stay on GitHub, where you can remove them.

Your choices

Disconnect any source in Settings, or revoke DoomScroll.build's access in your GitHub, Google or Microsoft account settings.
Turn on Stealth mode so DoomScroll.build never writes to GitHub.
Delete your account in Settings at any time.

Children

DoomScroll.build is a work tool and is not directed at children under 13.

Contact

Questions or requests: open an issue at github.com/doomscroll-build/doomscroll-plugin/issues.