Privacy Policy
Who We Are
We are Godspeed Systems LLC, a Limited Liability Company registered in the state of Oregon. We build software and services for day-to-day productivity. Submerse is one of our products/services.
Any "we," "our," or "us" in this policy refers to Godspeed Systems LLC. Any "you" in this policy refers to you, the user (whether a person or a business) of Submerse.
What This Is
This is the Privacy Policy for Submerse, a service that allows you to save and organize bookmarks. If you read the whole thing, you will understand how we collect, use, manage, and protect your information when you use our web application at https://submerse.it, our Browser Extension, our iOS application, our command-line tool, and our Android application once it is released (collectively, the "Service")
Read with care. By using Submerse, you agree to the collection and use of information in accordance with this policy.
Information We Collect
1. Personal Information You Provide
When you create an account or use our Service, we collect:
- Account Information: Your name and email address (collected and managed through our authentication provider, Clerk)
- Email Choices: Whether you get product updates and tips (see Communication). When you say yes, we keep a record of it with your account: when you said it, where (when you signed up, in Settings, or on Resend’s email preferences page), and, for a yes given in Submerse, the exact sentence you were shown. When you unsubscribe, we keep when. We also note when the Browser Extension first saves something new for you, so the email about the extension is not sent once it has.
- Bookmarks: URLs you save, text notes and voice notes you add as bookmarks of their own, custom titles, and personal notes you add
- Collections (Pools): Names and descriptions of bookmark collections you create
- Followed Pools: The public pools you follow. A pool’s owner sees how many people follow it, not who.
- Workspace Information: Your workspace’s name and settings
- Command-Line Access: If you sign in to the Submerse command-line tool, we keep a record of the key it uses: the name it was created with (which includes your computer's user name and host name), when it was created, and when it was last used. We store only a one-way hash of the key itself, never the key. You can see and revoke these keys at any time in Settings. When a new key is created, we email you its name and when it was created, so you can revoke it if it wasn’t you.
2. Automatically Collected Information
Web Application
- Usage Data: Information about how you interact with our Service
- Device Information: Browser type, operating system, and device identifiers
- Preferences: Your theme (a built-in theme or custom themes you create) and your Auto-enrich Bookmarks setting, stored with your account so they follow you across devices and cached in your browser so pages open in your theme. Other choices (list or grid view, grid size, the sidebar, paste-from-anywhere, and the last workspace and pool you opened) are kept only in your browser.
- Share Link Analytics: When someone opens an
sbmr.seshare link, we record the open — its timestamp, the link code, the referring page, and the browser user-agent — to provide aggregate analytics on shared pools. We do not log IP addresses for share links.
Browser Extension
The Browser Extension collects:
- Current Tab Information: The current page’s title and URL when you open the extension, which saves that page to your account unless it is already there or is a Submerse page
- Images You Save: When you right-click an image and choose “Save image to Submerse”, the image file, its address, the address of the page it came from, and the image’s description (its alt text) when the page gives one
- Local Cache Data: Your last-used workspace and your theme, kept in the extension’s storage in your browser. Versions before 0.5.0 also kept the address of each page you saved, with the notes and pools you chose for it; updating the extension deletes them.
| Permission | Purpose |
|---|---|
storage | To keep, in your browser, your sign-in session, your last-used workspace, your theme’s two colours so the toolbar icon can show your theme, and, while an image save is running, a marker so an interrupted save can be reported |
tabs | To read the current tab's title and URL when you open the extension, and save that page |
activeTab | To interact with the active browser tab |
cookies | For authentication session management |
contextMenus | To add “Save image to Submerse” to the menu when you right-click an image |
scripting | To download an image you chose to save, and read its description (alt text), from within the page it appears on, using that site’s cookies as the page itself would |
notifications | To tell you whether an image save worked |
The Browser Extension can also reach Submerse’s own services (submerse.it, clerk.submerse.it and Convex at *.convex.cloud) and our error and performance reporter, Sentry (*.sentry.io). The first time you save an image hosted on a different site from the page it appears on, the extension asks for access to all websites, which your browser describes as reading and changing your data on all websites. It uses that access only to download the images you choose to save, wherever they are hosted, and to read their descriptions (alt text). It is not granted at install. Once you allow it, it is not asked for again; if you decline, it is asked for again the next time an image needs it.
3. Content We Process
When you save a bookmark, we may automatically:
- Web Page Content: Scrape and store the page content in markdown format for the "read-it-later" feature
- Page Metadata: Extract and store the page title from the HTML
- AI-Generated Summaries: If you have auto-summarization enabled, we generate brief summaries of the page content using AI
- Uploaded Files: Images, PDFs and voice recordings you upload as bookmarks are stored on our CDN
- Voice Note Transcripts: When you upload a voice recording, we send the audio to ElevenLabs to be transcribed, and store the transcript as the text of the note
- Image Colors: We work out the main colors of each image you save, on our own servers, whatever your auto-summarization setting, so search can find it by color. The image is sent to no other service for this; the words naming its colors go to Turbopuffer, our search index (see below)
- Image Materials and Textures: If you have auto-summarization enabled, we send a copy of each image you save to TypeSafe, at most 1568 pixels on its long side and without its location or other metadata, and keep the materials, textures, finishes and patterns its Jev model is very sure the image shows, so search can find it by them. The words go to Turbopuffer, our search index (see below), and are never shown. An image you save while auto-summarization is off is not sent, then or later, unless you crop it, replace it or restore its original once it is back on. For images saved before we started, we go by your setting at the time we get to them, and skip any that our records show were saved with auto-summarization off. Words already found stay in search until you delete the image
You can disable automatic summarization at any time in your account settings.
How We Use Your Information
We use the collected information to:
1. Provide and Maintain the Service
- Store and organize your bookmarks
- Sync your data across devices
- Enable collaboration features (pool sharing)
2. Enhance User Experience
- Generate AI summaries of bookmarked pages (if enabled)
- Suggest starter pools from the interests you type on the Onboarding screen
- Provide search across your bookmarks’ titles, URLs and notes (including AI summaries and voice-note transcripts), the main colors of your images, and the materials, textures, finishes and patterns found in them while auto-summarization was enabled
- Remember your preferences and settings
3. Communication
- Send email notifications when someone shares a pool with you
- Send security emails about your account, such as when a new command-line access key is created
- Send service-related announcements (rare and important only)
- Send product updates and tips, only if you are subscribed to them: what’s new in Submerse, tips for using it, and, a few days after you sign up, one email about the Browser Extension if you haven’t saved anything new with it yet
Product updates and tips are opt-in. Right after you make your account on the website, we ask once, with a checkbox left unticked, separate from agreeing to the Terms. If you make your account in our iOS or Android app, you aren’t subscribed until you turn on “Product updates and tips” in Settings → Profile on the website. Accounts that already existed when we started sending these emails (all of them in the United States) were subscribed, and can unsubscribe at any time.
Every one of these emails has an unsubscribe link, and you can turn them off in Settings whenever you like. Unsubscribing from them stops nothing else: pool invitations, billing emails and security emails are always sent, and emails about activity in pools you share keep their own switch.
4. Service Improvement
- Analyze usage patterns to improve features
- Debug and resolve technical issues
Third-Party Services
We use the following third-party services to operate Submerse:
Essential Services
Clerk
Purpose: User authentication (including a check, when you sign up, that you’re a person and not a bot), account management, and subscription checkout and billing
Data Shared: Email address, name, for accounts made before we moved that question off the sign-up page, the answer to its product updates and tips checkbox, and, if you subscribe to a plan, the payment details you enter at checkout and any promo code you use. Those go to Clerk and its payment processor. We never see your full card number, and we don’t store payment details. When you sign up, Clerk runs a Cloudflare Turnstile check in your browser, which shares technical details about your browser and device with Cloudflare.
Privacy Policy: clerk.com/legal/privacy
Convex
Purpose: Backend database and real-time synchronization
Data Shared: All user content (bookmarks, pools, workspaces)
Privacy Policy: convex.dev/legal/privacy
Turbopuffer
Purpose: Search index for your bookmarks
Data Shared: Each bookmark’s title, notes (your own, AI summaries and voice-note transcripts) and URL, the identifiers of the bookmark and its workspace, owner and pools, and whether it is in the trash; for an image, words naming its main colors, including broader names such as blue for navy, and the materials, textures, finishes and patterns found in it while auto-summarization was enabled; and the words you search for
When Used: Whenever a bookmark or its pool membership is saved or changed, when we work out an image’s colors or its materials, textures, finishes and patterns (including for images you saved before we started), and when you search
Privacy Policy: turbopuffer.com/privacy-policy
Vercel
Purpose: Web application hosting
Data Shared: Connection metadata, usage analytics
Privacy Policy: vercel.com/legal/privacy-policy
Context.dev
Purpose: Web page content extraction and conversion to markdown
Data Shared: URLs you bookmark
When Used: Only when you save a bookmark
Privacy Policy: context.dev/privacy
Anthropic
Purpose: AI-powered summaries of the pages, images and PDFs you save, cleanup of page text our own cleaning cannot tidy, and suggested starter pools from the interests you type on the Onboarding screen, using Claude
Data Shared: Page content (up to 15,000 characters) with the page’s title and address; images you save as bookmarks; the first two pages of PDFs you save as bookmarks (the whole file, if it cannot be split); with each summary request, the names of the labels already in the bookmark’s workspace (including labels other members created) and any enrichment prompts you have written in your settings; and the interests you type on the Onboarding screen
When Used: Summaries, labels and cleanup: only if auto-summarization is enabled in your settings. Pool suggestions: each time you submit your interests on the Onboarding screen, whatever that setting is
Privacy Policy: anthropic.com/privacy
TypeSafe
Status: In use for the images you save. Its use for pages is planned and not yet in use: no page you save is sent to TypeSafe today.
Purpose: Naming the materials, textures, finishes and patterns in the images you save, so search can find them by those words. Planned: deciding which parts of a saved page are its main content
Data Shared: A copy of each image you save, at most 1568 pixels on its long side and without its location or other metadata, with fixed questions about what it shows. Planned: a saved page’s cleaned text, split into sections and sent a few sections at a time, with the page’s title and address
When Used: Only if auto-summarization is enabled in your settings: when you save an image, and again when you crop it, replace it or restore its original; for images you saved before we started; and again when we change the questions we ask, how we read the answers, or the model we ask. When the planned use begins, only for pages whose text our own cleaning cannot tidy
Privacy Policy: typesafe.ai/legal/privacy-policy
ElevenLabs
Purpose: Transcribing the voice notes you upload (speech to text)
Data Shared: The audio file of each voice note, sent from our servers. The transcript that comes back becomes the text of your note.
When Used: Every time you upload audio, whatever auto-summarization is set to in your settings
Retention: Submerse asks ElevenLabs to delete the recording and transcript it holds for each voice note once the transcript has come back. If that request fails, or ElevenLabs does not answer it, its copy stays under ElevenLabs’ own retention (see its privacy policy).
Privacy Policy: elevenlabs.io/privacy-policy
Resend
Purpose: Transactional and notification email delivery, and product updates and tips for people who are subscribed to them
Data Shared: Email addresses and user's first name, and what each email says — for the email about a new command-line access key, that includes the key’s name, which contains your computer’s user name and host name; for the digest of what was added to a pool you share, it includes the titles of what was added (a voice note’s title can be the opening words of its transcript). Resend also keeps a contact for each account, with its email address, first name, and whether it is subscribed to product updates and tips, so those emails go only to people who are. If you unsubscribe, or subscribe again, on Resend’s own email preferences page, Resend tells us, and your setting in Submerse changes to match.
When Used: When a notification-worthy action is performed (for example, sharing a pool with someone); when we send product updates and tips; and when your contact has to change: when you sign up, change your email address, subscribe or unsubscribe, or delete your account, which deletes it. Accounts that existed when we started were each added once.
Privacy Policy: resend.com/legal/privacy-policy
Sentry
Purpose: Error tracking, performance monitoring, and the “Report a Bug” form
Data Shared: Error messages, stack traces, and performance metrics, with technical details about the session in which the error happened — the page you were on and the page you came from (their addresses, which include the names of the pool and bookmark you had open and can include search terms), your browser or device and its operating system version, your language and time zone, your browser window size, the identifier of your account, which workspace and pool were selected, your theme, whether your sign-in was working, the state of your bookmark list (for example, whether it had loaded), how long the page had been open, and a short record of the pages you visited, the buttons you pressed and the requests the page made just before the error — so we can tell what went wrong without asking you. When we measure how quickly a page loads, the measurement carries the page’s address without its search terms. Our apps also send short technical logs of connection and request activity, such as a request that had to be retried. While you are signed in, an error report also carries a link to your PostHog session, which shows its replay if one was recorded. When you submit a bug report, we additionally send what you wrote and the email address and name on your account, so we can reply.
When Used: When errors occur in the application, when we measure how quickly pages load, and when you choose to submit a bug report
Privacy Policy: sentry.io/privacy
PostHog
Purpose: Product analytics to understand how features are used
Data Shared: Page views, feature usage events, email address, name, and workspace association, and, when Submerse processes a bookmark you saved, that bookmark’s address and your account identifier
When Used: While you are signed in and using the Service, and whenever Submerse processes a bookmark you saved, from any app
Privacy Policy: posthog.com/privacy
Jam
Purpose: Bug reporting — capturing a short screen recording plus technical diagnostics when you choose to report a problem
Data Shared: A screen recording of the Submerse tab, browser console logs, network requests and responses made by the page, the actions you took during the recording, your browser and device details, and technical context such as your user and workspace identifiers, the email address on your account and the page you were on
When Used: Only when a recording is explicitly started and then submitted. Jam records nothing in the background: starting a recording and sending it are two separate actions, and you can decline either. If you never start a recording, nothing is captured or sent.
Privacy Policy: jam.dev/privacy
YouTube Data API (Google)
Purpose: Enriching YouTube bookmark metadata (title, description)
Data Shared: YouTube video IDs from URLs you bookmark
When Used: Automatically when you bookmark a YouTube video
Privacy Policy: policies.google.com/privacy
Other sites’ public APIs
Purpose: Fetching a saved link’s public details (such as its title, description or summary text) from the site it came from
Data Shared: Only the link or its identifier (for example, a repository name, a question number or an article title)
When Used: Automatically when you save a link from GitHub, Stack Overflow or another Stack Exchange site, Hacker News, Wikipedia or YouTube (GitHub API, Stack Exchange API, Hacker News API, Wikipedia REST API, YouTube oEmbed, and the public pages of those sites)
Privacy Policy: Each site’s own
Bunny.net
Purpose: CDN hosting for uploaded files (images, PDFs, voice recordings)
Data Shared: Files you upload as bookmarks
When Used: When you upload a file as a bookmark
Privacy Policy: bunny.net/privacy
Screen recording and session replay
Two of the services above can capture what is on your screen inside Submerse, rather than only recording that an event happened. We treat these separately because they collect a different kind of data:
- Bug reports you submit (Jam). Starting a recording is always an explicit action, and the recording is only sent if you then choose to submit it. A submitted recording can include your bookmarks, pool names, notes and other content that was on screen at the time, along with the technical diagnostics listed above.
- Session replay (PostHog). Where enabled, this reconstructs how you moved through the app so we can diagnose problems that are otherwise impossible to reproduce. It runs only while you are signed in.
Both services filter credentials — authentication tokens, passwords and payment fields — on your device before anything is uploaded. That filtering is automated and we do not treat it as a guarantee: please avoid entering sensitive information you would not want captured while a recording is running.
Recordings and replays are used only to diagnose and fix problems. They are never used for advertising or sold, and we do not use them to build a profile of you.
Data Storage and Security
Storage Location
- Your data is stored securely in Convex's cloud infrastructure
- Uploaded files (images, PDFs, voice recordings) are stored on Bunny.net's CDN infrastructure
- Local browser storage (localStorage, cookies and chrome.storage) holds your sign-in session, your preferences, and, once you have signed in to the web app, our analytics provider PostHog’s identifiers for your browser and session, as described above
Security Measures
We implement industry-standard security measures including:
- Encrypted data transmission (HTTPS/TLS)
- Secure authentication tokens
- Minimal permission requirements for the Browser Extension
Data Retention
- Active account data is retained as long as your account remains active
- Bookmarks you move to the Trash stay there until you empty it (or delete your account), and are then permanently deleted
- You can delete your account, and everything in it, at any time from Settings
Your Data Rights
You have the following rights regarding your data:
Access and Control
- View Your Data: Access all your bookmarks, pools, and account information through the Service
- Export Your Data: Request a copy of your data in a portable format
- Delete Your Data: Delete individual bookmarks, pools, or your entire account
- Modify Your Data: Edit or update your bookmarks, notes, and account information at any time
Privacy Controls
- Disable AI Summarization: Turn off automatic page summarization in Settings
- Private Pools: Create private bookmark collections visible only to you
- Sharing Controls: Control who can access your shared pools
- Product Updates and Tips: Turn them on or off with “Product updates and tips” in Settings → Profile on the website, or unsubscribe with the link in any of these emails. Pool invitations, billing emails and security emails are still sent
Account Deletion
To delete your account and all associated data:
- Go to Settings in the web application
- Under "Delete Account", click "Delete Everything"
- Type your email address to confirm
In the iOS app, the path is Settings → Delete Account: type DELETE, then tap "Delete my account". (The Android app will offer the same path when it is released.)
You can also email support@submerse.it and we will delete your account for you.
Account deletion is permanent and cannot be undone. It starts immediately and usually finishes within minutes. We permanently delete your bookmarks and notes, your pools (people you shared them with, and anyone following them, lose access), the record of which pools you follow, your labels, the files you uploaded, the page copies we saved for reading and search, your settings and email choices, your command-line access keys, your contact with our email provider, Resend, and your sign-in account with our authentication provider.
A few things are not removed at that moment:
- Backups: our database backups keep a copy for up to 7 days, after which it expires
- Emails we sent: we keep copies of the emails we send (including your account-deletion confirmation) for up to 30 days, and our email provider keeps delivery logs under its own retention period
- CDN caches: deleted files can still be served from our CDN's edge caches for up to 30 days, until those caches expire
- Other people's bookmarks: bookmarks that belong to someone else stay in their pools, including ones you added to a pool or labelled there — they are theirs, and they remain without your name on them
- Accounting records: if you took part in our affiliate program, commission and payout records we are required to keep for tax purposes are retained, without your account
- Data already held by third parties: information shared with the services listed above (for example, an error report or a bug-report recording) is governed by their retention periods
Browser Extension Privacy
Regarding the Submerse Browser Extension:
What It Accesses
- Accesses a tab only when you act: opening the extension saves the current tab’s page (its title and URL) to your account, unless it is already there or is a Submerse page; choosing “Save image to Submerse” on an image reads that image (downloaded from within the page you are on, or directly by the extension if that fails), its address, and the page’s title and address; the first time an image is hosted on a different site from its page, it asks for access to all websites, which it uses only for these downloads and descriptions
- Requires authentication through the main web application
- Cannot be used without an active Submerse account
What It Stores
- Stores authentication tokens securely via Clerk
- Stores your last-used workspace preference
- Keeps your Submerse theme in the extension’s own storage so it opens in your theme and its toolbar icon shows it
- While an image save is running, keeps a temporary marker so an interrupted save can be reported
What It Does NOT Do
- Does not track your browsing history
- Does not read the content of the pages you visit. The page you open the extension on is saved to your account (unless it is already there, or is a Submerse page), and our servers then fetch that page’s content, and summarize it if auto-summarization is on, as for any bookmark you save. The one thing it reads from a page itself is the description (alt text) of an image you choose to save with “Save image to Submerse”, as described above
- Does not collect data from inactive or background tabs
- Does not share data with third parties except as necessary to save bookmarks and to send error reports and performance timings to Sentry (as described in this policy)
Contact Us
If you have questions, concerns, or requests regarding this Privacy Policy or our data practices, please contact us at:
Email: support@submerse.it
Changes to This Privacy Policy
We may update this Privacy Policy from time to time to reflect changes in our practices or for legal, operational, or regulatory reasons. We will notify you of any material changes by posting the new Privacy Policy on this page and updating the "Last Updated" date at the top.
Your continued use of the Service after changes become effective constitutes acceptance of the updated Privacy Policy.