Capter is run by Muneeb Ahmad, an individual developer in Pakistan. He decides how your data is used and is responsible for it. Where this policy says "we", it means him. Capter's whole job is reading your screenshots, so you deserve a straight answer about what it collects and where that goes. If anything is unclear, email hello@capter.app and he'll explain.
1. What we collect
A few kinds of data, and that's the lot.
- Your account
- The email and name you sign up with, your password (stored hashed, so we can't read it), a profile picture if you add one, and your time zone, so deadlines land on the right day. If you sign in with Google or Apple instead, that provider gives us the name and email you choose to share. Apple may provide a private relay address instead of your real email.
- Your captures
- The screenshots, images and files you choose, the note you add, and the tasks and tags they turn into. A capture reaches us only when you take one with the desktop shortcut, drop or paste an image or file into Capter, or type a task. In the phone apps, it also reaches us when you pick, take or share an image there. We can't see your screen or photo library any other way.
- Where a capture came from
- With each capture from the desktop app, Capter records the app you captured from, its window title and, for a browser, the page address. The task shows them as where it came from, and they go to the AI and to the tools you connect (sections 5 and 9). For a window Capter recognises as private or incognito, it skips the window title and the page address.
- Your connections
- If you link Google Calendar, Outlook Calendar, Google Tasks, Todoist, TickTick, Microsoft To Do, Jira, Linear, Asana, ClickUp, monday.com or Notion, we store the access tokens those services hand us, encrypted, so Capter can add an event or a task, or file an issue, when you ask it to. For a tool with projects or lists, we also keep the names of your projects, lists, teams, boards or databases there, so Capter can file each task in the right one. Where the service tells us, we also keep the email address and ID of the account you linked (for Notion, the workspace's name), so Settings can show which account is connected. For Jira, Atlassian requires us to tell it once a week which of its accounts we hold; when it says one was closed, we delete that account's details. Disconnect any time and the tokens and account details are deleted.
- Your plan
- If you buy Pro, Paddle, which runs our checkout, tells us your Paddle customer and subscription IDs, your plan and billing interval, its renewal date and whether a payment went through, so your account gets Pro. Your payment details go to Paddle only; we never see them.
- Notifications
- When you use the phone apps and allow notifications, we store the device's push token with your account so we can deliver the reminders you enable. Signing out removes that device token.
- Operational logs
- Enough to tell when something breaks: a failed request, a crash. Crash and error reports from the desktop apps include a pseudonymous identifier derived from your account while you're signed in. Error reports from our servers can include your Capter account ID. Neither kind carries your email or name on purpose. We don't run ad trackers.
2. What we don't collect
Capter never watches your screen in the background and never scans your photo library. It reads your clipboard only when you paste into it. Nothing about your screen or files leaves your device until you capture, paste, drop or share something. Besides your own tasks and settings, the desktop apps send crash reports, and the Mac app checks Capter's server for updates. Crash reports from the desktop app carry technical details (app version, OS version, what the app was doing when it fell over), where in Capter's code it crashed and the names of the program files it had loaded, and, while you are signed in, a pseudonymous account-derived identifier. They never include the app's memory, your screenshots or your files.
We also don't sell your data.
3. How your screenshots are used
Your captures exist so the product can work for you. The image goes to Google's Gemini API, along with the details listed in section 9, and the AI drafts a task title, a deadline and tags. After that, the image and the task stay in your account until you delete them. No other Capter user can see them. Capter doesn't train AI models on them, and under Google's terms for the paid Gemini API, Google doesn't either. If you send a task to a tool you've connected, such as Jira or Todoist, that tool keeps its copy under its own terms. The only person who could look at your captures is Muneeb, who runs Capter, and he only does that when you ask for help; section 6 explains how.
4. Why we're allowed to hold this
If you're in the UK or EU, the law asks us to name why we're allowed to hold each piece of data. This is the legal basis for each.
To run the service you signed up for: your account, your captures and where they came from, the tasks they become, the AI draft we make the moment you capture, the tasks you send to tools you've connected, and the reminder emails. Reminders are on by default because reminding you is the product's job, and the unsubscribe link in any reminder email stops them. This is us performing our contract with you (GDPR Article 6(1)(b)). If you buy Pro, keeping your plan and payment status rests on the same contract.
To keep the service working and safe: sign-in security, the bot checks on our forms, and crash and error reports. This is our legitimate interest in a product that stays up and isn't abused (Article 6(1)(f)). Reports may be connected across failures with the identifiers described above, but we do not use them for advertising or behavioral profiling.
To hear how Capter is working for you: we send one email a few days after you sign up to ask how it's going (the link in it unsubscribes you), and we keep the feedback you send from the app, including a quick rating, with your account so we can improve Capter. This rests on our legitimate interest in knowing whether Capter works for the people who use it (Article 6(1)(f)).
AI personalization rests on that same legitimate interest, which is exactly why it ships with an off switch and a one-click erase (section 9). Turn it off and we stop.
We don't run anything that needs a separate consent banner, because we don't do advertising, cross-site tracking, or profiling.
5. Who else touches your data
Running Capter takes a few outside services. Each one below handles your data on Capter's behalf, under data-processing terms that come with its service, except where the row says otherwise.
- Cloudflare
- Hosts this site, the web app and the API, and stores your screenshots and files in Cloudflare R2.
- Neon
- Hosts the database that holds your account, tasks and settings.
- Google (Gemini API)
- Reads what you send Capter to turn into a task and drafts it (section 9).
- Resend
- Sends sign-in, reminder and account emails, and passes on messages you send through our contact form. A reminder's subject and body include the task's title.
- Sentry
- Collects crash and error reports from the desktop apps and from our servers.
- Expo
- When you use the phone apps, delivers your reminders to your phone. Expo hands each one to Apple's or Google's push service (APNs or FCM), and the notification includes the task's title.
- Cloudflare Turnstile
- Checks for bots on our sign-in, sign-up, password-reset and contact forms. Cloudflare also uses the signals it collects there to run and improve Turnstile, and for that it decides how they're used itself, under its Turnstile privacy policy.
Payments. If you buy Pro, you buy it from Paddle.com, our reseller and merchant of record. Paddle collects your payment details, billing address and any tax information itself and handles them under its own privacy notice, shown at checkout. When you open checkout we pass Paddle your account email, so you don't have to type it in. Paddle tells us only what we need to give your account Pro, listed under "Your plan" in section 1.
Tools you connect. When a task goes to Google Calendar, Outlook Calendar, Google Tasks, Todoist, TickTick, Microsoft To Do, Jira, Linear, Asana, ClickUp, monday.com or Notion, whether you pick the tool or let Capter route it, that tool receives the task's title, description and deadline, a link back to the task in Capter, and where it was captured: the app, the window title and the page address. The tools get the page address without its query string or fragment; Capter itself keeps the full address. Jira, Todoist, Asana, ClickUp and Microsoft To Do also receive the screenshot and attached files. Jira and Todoist receive your tags as labels, and TickTick receives your tags. The task then sits in your account with that service, under its own terms and privacy policy.
Where this happens. Capter's database is with Neon in Singapore (AWS's ap-southeast-1 region), and the API runs next to it in Singapore. Screenshots and files are in Cloudflare R2, which doesn't keep them in one fixed country. Sentry keeps error reports in its EU region. Several of these providers are US companies and may process your data in the United States. Capter itself is run from Pakistan. Where any of this counts as a transfer out of the UK or EU, each provider's data-processing terms include the EU Standard Contractual Clauses, plus the UK addendum to them where the provider offers one. Google's Gemini API comes under Google's data processing addendum for processors, which uses the clauses as a fallback.
We have never received a government request for user data. If we ever do, we'll push back on anything overly broad and update this page.
6. Security and team access
Everything travels over TLS and is encrypted where it's stored. Passwords are hashed. Tokens for any service you connect are encrypted with AES-256-GCM. A web sign-in ends after a week without use; the desktop and phone apps use revocable device tokens that renew as you use them and lapse after long inactivity (a year at the outside).
Capter isn't end-to-end encrypted. In practice that means Muneeb, who runs Capter, can technically access your account. He does so only when you ask, usually a support request where he needs to see the task or screenshot you're writing in about. If you never reach out, he never looks. The exceptions are clear abuse of the service and valid legal orders he's required to answer.
7. Your rights
You can delete any screenshot or task whenever you like, and deletion is permanent. You can also permanently delete your whole account from Settings in the web app or the phone apps; our account-deletion page explains exactly where. Your AI data (preferences, memory, feedback) has its own download and erase buttons in web settings. For a full copy of your data or any help using these controls, email hello@capter.app; a person will handle it, normally within a few days and always within a month.
Deleting your account removes your data from our live database and storage; backup copies then age out within 30 days. A few records stay on purpose. If you were a founding member, we keep your email address in a normalized form (lowercase, without any +tag) so the same address can't claim a second founding place. If email to your address bounced or was marked as spam, we keep the address on a do-not-email list. The IDs of any Paddle purchases stay in our billing records without a link to your account, and Paddle keeps its own records of them. Crash and error reports and email delivery logs at Sentry and Resend expire on those services' own schedules.
If you're in the UK or EU, the GDPR gives you the right to access your data, correct it, delete it, export it, or ask us to restrict or stop a particular use. Some of that is self-serve today (delete a capture or account, export or wipe your AI data); for the rest, email us and we'll do it by hand. You also have the right to complain to your local data-protection authority if you think we've mishandled something, though we'd rather you gave us the chance to put it right first.
If you're in California, the CCPA gives you the matching rights to know what we hold, delete it, and correct it, plus the right not to be treated differently for using them. The categories we hold are identifiers (your name and email), the content you create (screenshots, files, notes, tasks, tags), where each capture came from (the app, window title and page address), and basic technical logs. We do not sell your personal information and we do not share it for advertising. There's no opt-out to click because there's nothing to opt out of.
8. Cookies
Capter sets cookies only to keep you signed in, to finish a desktop sign-in that you complete in the browser, and to carry your time zone through a Google or Apple sign-in. A Google or Apple sign-in also sets a short-lived security cookie. None of them track you across the web or feed an ad network. The Turnstile bot check and Paddle's checkout run Cloudflare's and Paddle's own code, which their own privacy notices cover.
9. AI drafting and personalization
When you capture something or type a task, Capter sends it to Google's Gemini API so the AI can suggest a title, a deadline and tags. Along with the image or the line you typed, the AI receives:
- your note, and the text of an attached PDF or text file when your note asks it to read one
- the app, window title and page address the capture came from
- your name, your time zone, and the AI preferences and memory you've saved
- your tags' names and descriptions, with a few recent task titles from each tag
- if Capter routes tasks for you: the names of your Jira projects and connected tools, your routing rules, and recent tasks where you changed or removed the Jira project Capter picked
- if you've connected Todoist, TickTick, Microsoft To Do, Linear, Asana, ClickUp, monday.com or Notion: the names of your projects, lists, teams, boards or databases there, so it can pick where each task goes
Under Google's terms for the paid Gemini API, Google doesn't use any of this to train its models. Google does log prompts and responses for a limited time to detect abuse, and for safety, security and legal reasons. Capter has AI Studio's optional storage of requests turned off. Section 5 covers where this processing happens.
If you leave personalization on, Capter also keeps short notes about how you like tasks written, learned from your edits and from Give feedback on this draft in a task's More menu. A note looks like "prefers short titles". Your edits and feedback go to the Gemini API to write these notes. A note is a short summary written by the model and doesn't copy your tasks.
Settings gives you the full kit: read every note, download them as a file, delete one, wipe them all, or switch personalization off entirely. Notes stay until you delete them. Turning the feature off freezes them without deleting, so you can still export first. The AI personalization page walks through all of it.
Google user data stays out of the AI. Data Capter receives from Google APIs (your Google Calendar events or Google Tasks, for example) is used only to do the thing you asked for, such as adding an event or a task, and to show you the result. It is never sent to any AI or machine-learning service, never used to train or improve any model, and never sold or passed to anyone else. The AI only sees what you send Capter to turn into a task, with the details listed above. Capter's use and transfer of information received from Google APIs to any other app will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
10. Children
Capter isn't meant for children. You need to be at least 13 to use it, and older where your country sets a higher age for digital consent (16 in parts of the EU). We don't knowingly collect anything from a child under that age, and if we find out we have, we delete it. If you believe a child has given us their data, email hello@capter.app and we'll sort it.
11. Changes to this policy
Products change and this policy will too. If a change matters, we'll email registered users at least seven days before it takes effect. The date at the top of this page is always the version you're reading.
Questions, or anything that reads as wrong? Email us and a human will reply.