Privacy Policy
Last updated November 11, 2026
This policy explains what personal data ShipVitals collects, why, and what you can do about it. ShipVitals is run by Nanner Tech LLC ("we", "us"). Questions go to support@nannertech.com.
Who is responsible
- For your account (who you are, how you sign in, your preferences), Nanner Tech LLC is the controller.
- For the data in a workspace (connections, figures, projects, people), the workspace's owner decides what goes in, and we process it for them. The Data Processing Addendum covers that.
What we collect
Your account: your email address; your name and profile picture, as your sign-in provider (Google or GitHub) shares them, and we keep nothing else it shares; which sign-in methods you use; your theme, number and date format, time zone, first day of the week and keyboard shortcut choices; the version of the Terms you accepted and when; and whether two-factor sign-in is on.
Security records: security changes to your account (a sign-in method added or removed, two-factor sign-in turned on or off, a recovery code used, every device signed out, an email change), with the time and, for most, your browser and a hashed IP address; an email change also records the old and new addresses. Recovery codes are stored only as keyed hashes.
Workspace data: the credentials for the services a workspace connects, encrypted with that workspace's own key; the figures ShipVitals reads from them (costs, usage, revenue totals, test and uptime results); projects; members, their roles and the email addresses invited; and the workspace's audit log.
Devices you sign in: when you sign in the desktop app or vitals, its name, operating system and version, and when it was last used. See The desktop app and the vitals runner.
Billing: for a workspace on a paid plan, its billing email, its plan, add-ons and their dates, and Stripe's ids for the customer and the subscription. See Paying for ShipVitals.
What a workspace adds: bills forwarded to its invoice inbox; the sites, test accounts and results of its cloud browser checks and guide walks; its API tokens; and the feedback and support requests its members send.
What we never collect: your customers' names, emails or card details (revenue is counted against a hashed customer id), your source code, or your users' identities. ShipVitals has no advertising, no third-party trackers and no third-party analytics scripts; our own analytics are described under Analytics.
Why we use it
- To run ShipVitals: sign you in, show each workspace its figures, enforce who sees what, and keep it secure.
- To send emails the service needs, such as sign-in links, invitations, security notices, exports and deletion notices. We don't send marketing email.
- To understand how the service is used, from server-side counts that carry no personal data (for example, how many sign-ins completed, how long a new machine took to send its first result, or why checks couldn't run). These counts are keyed by a one-way hash of the workspace and hold only fixed codes and durations: never a command, a file path, a test's name, a machine's name or an amount.
- To see whether ShipVitals works for the people who try it, from cookieless page counts: see Analytics.
- To meet legal obligations and to protect ShipVitals and its users from abuse.
Where the GDPR applies, our legal bases are the contract with you (running the service), our legitimate interests (security, abuse prevention and improving the service) and legal obligations.
Analytics
ShipVitals counts how its pages are used with Umami, an open-source analytics tool that we run ourselves. It sets no cookies and stores nothing on your device.
- What is sent: a page view for each page you open, and these steps: opening the demo and its tabs; starting a sign-up and finishing it; being shown a plan limit (by its name, such as projects); opening the plan sheet; starting Checkout (with the plan and whether monthly or yearly) and the subscription becoming active (with the plan); creating an API token; testing the MCP server (with whether it worked); and asking to hear when a service is ready (with the service's name). On the public site the page is its address. In the app it is only the kind of page, such as
/app/:slug/projects/:id: never a workspace's or project's name or id, a query string or anything on the page. If you came from another site, that site's address without any path is sent with your first page. - How it is sent: your browser sends it to ShipVitals' own address, never to a third party, and our servers pass it on to Umami. Before they do, they cut your IP address to its network (an IPv4 address's last number and an IPv6 address's last 80 bits are removed) and your browser's identification to its name alone (such as Firefox or Safari, with no version, system or device), and they drop visits from robots. Nothing else about you, your device or your account goes with it.
- What we do with it: weekly counts of visitors, sign-ups, first numbers, first profits and paid plans, and of how often connections sync without failing. Umami can't tell apart people on the same network using the same browser, and we don't try to.
Our legal basis is our legitimate interest in knowing whether ShipVitals works for the people who try it.
Paying for ShipVitals
Paid plans are paid through Stripe. When a workspace's owner chooses a plan, ShipVitals sends Stripe the billing email (the owner's, unless another has been set), the workspace's name and id, and the plan, and then sends the owner to Stripe's own checkout page. Card details, the billing address and any tax ID are entered there and kept by Stripe, under its own privacy policy: ShipVitals never receives a full card number. Stripe tells us back what we need to run the plan: the subscription, its period and state, and whether a payment went through. We show the workspace's last invoices from Stripe to its owner and admins, without keeping them. Our legal basis is the contract with the workspace's owner.
The invoice inbox
Each workspace can have its own address to forward bills to. A forwarded mail reaches Mailgun, which receives it for us, and then ShipVitals.
- What we keep: the sender's address, the subject, the date, the bill itself (the attachment, or the mail's text, never its HTML) and what was read from it: the vendor, the period, the total, the currency and the tax. Bills' files are deleted after your plan's evidence period (7, 30 or 90 days); what was read from them stays with the workspace's costs. A mail marked "Not an invoice" is deleted with its file 7 days later.
- Whose mail is read: ShipVitals checks each mail's DKIM signature itself, against the sender's domain. Only a mail that proves it comes from a sender the workspace has confirmed is read automatically; any other waits, unread, until an owner or admin confirms its sender or, for a mail that fails the check, chooses to read it.
- How it is read: a bill is read by an AI model, as described under AI features.
Forward only bills you are allowed to share with us. They may contain the names and addresses on your own accounts with those vendors.
API tokens and the MCP server
A workspace's owner, admins and editors can make API tokens for scripts and AI agents, which read the workspace's figures through ShipVitals' API or its MCP server. We store a one-way hash of each token, never the token itself, with its name, who made it, the projects it covers, when it expires, and when it was last used and by which tool (the name the tool gives, such as an AI assistant's). A token can only read, and only what its maker can still see.
Cloud browser checks and guide walks
Cloud browser checks open a page of your site in a browser on our servers on a schedule. We keep the address and the conditions you set, each run's result and timings, and a screenshot of the page, which is deleted after your plan's evidence period.
Cloud guide walks follow a project's how-to guides on a test copy of your site, in a browser on our servers. ShipVitals reads the guides from the project's connected GitHub repository or its docs page. A walk can sign in as a test account you add to the project: its password is stored encrypted with the workspace's key, typed into your site by the browser, and never shown to the AI model. A walk's report, screenshots and findings are kept like a run's evidence, for your plan's evidence period.
The desktop app and the vitals runner
The ShipVitals desktop app and the vitals command line run your project's own checks on your machine. What they send is decided on your machine, by rules you can read in the app under This Mac → Privacy, and the first upload for each project shows you the exact data before it goes.
Sent to ShipVitals by default: each check's name, result, counts and duration; the commit, the branch and whether the working tree was clean; tool names and versions, the operating system and the machine's name; signing facts, file hashes and vulnerability counts; and, while a project's Failure details setting is on (it is by default), the names of failing tests and a short first error of at most 2 KB, with expected and received values removed. Turn Failure details off and only counts are sent.
Kept on your machine unless you choose otherwise: full logs, screenshots, test reports and traces. A project's Screenshots and Reports and traces settings, both off by default, send them so your teammates can see them; Share this failure sends one failure's evidence after showing you what it contains. What is sent this way is kept for your plan's evidence period (7, 30 or 90 days) and then deleted. Videos always stay on your machine: no setting sends them.
Never sent: folder paths and the contents of your files, environment values and secrets, keychain items and signing keys. In text that is sent, a file appears only as its name and line (for example checkout.spec.ts:12). All text passes through one scrubber on your machine first, which replaces values it recognises, such as the check's own secrets, key-shaped strings, Authorization and cookie headers, home folders and email addresses, with [hidden]. It can only hide what it recognises, so error messages can still contain what your app showed; keep screenshots off for projects with real customer data. Commands never cross the network: ShipVitals names a check, and your machine runs only checks a person approved on it.
Walking the guides: when you walk a project's guides, the desktop app calls an AI model through OpenRouter with your own OpenRouter key, which stays in your machine's keychain and is never sent to ShipVitals. Those calls go from your machine to OpenRouter, under your own agreement with OpenRouter, not through ShipVitals, and ask OpenRouter to route only to providers that keep no data and don't collect it (zero data retention). The model receives the guide's text, the test persona's name and role, the address of the site it walks and, for each step, the page's structure and a low-detail screenshot of the site the walk is allowed to open. What a walk costs is shown only on your machine. Confirmed findings are sent to ShipVitals as a check's failure details are: scrubbed, with paths cut to names, and their text only while Failure details is on. With it off, only each finding's id, severity, category, guide and step and its screenshots' file names are sent. The screenshots themselves stay on your machine.
Running a connection through a machine: when a connection runs through one of your always-on machines, its sign-in stays in that machine's keychain, and only what it reads is sent to ShipVitals: the service's sites (their names and ids) and their figures, how far it has read, and, when a sync fails, a scrubbed reason.
Devices and runners: signing in a device records its name, system and version, and we email you when it happens. You can see and revoke your devices in Account settings → Devices, and a workspace's admins can remove its machines in Settings → Runners.
AI features
ShipVitals uses AI models in five places. None of them receives credentials, raw billing exports or the contents of your files beyond what is listed here.
- Suggestions for repositories (when AI features are turned on for the service): for a connected repository that ShipVitals' own rules can't match to a project or a service, our servers send a model the repository's name, what the scan found in it (such as the names of the environment variables it uses) and the first 2 KB of its README, and ask for a project name and the services it seems to use. A person sees the suggestion as an editable prefill and decides.
- The digest's summary line (when AI features are turned on for the service): our servers send a model the facts in your own digest email, its subject and rows as you would read them, amounts you can see included, and ask for one sentence to put at the top. A line that doesn't match the facts is dropped, and the digest goes without it.
- Reading forwarded bills (when AI features are turned on for the service): for a bill from a confirmed sender, or one a member chose to read, our servers send a model the sender's address, the subject and the bill: its text (from the mail or the PDF, at most 24,000 characters), or the image when the bill is a picture. The model answers whether it is a bill, and its vendor, period, total, currency and tax. A reading ShipVitals isn't sure of waits for a member to check it.
- Cloud guide walks (when AI features are turned on for the service): our servers send a model the guide's text, the test account's name and role, the address of the site being walked and, for each step, the page's structure and a low-detail screenshot, as the desktop app's walks do.
- Walking the guides in the desktop app, on your own OpenRouter key, from your machine, as described under The desktop app and the vitals runner.
The first four go from our servers through OpenRouter and ask OpenRouter to route them only to providers that keep no data and don't collect it or train on it (zero data retention). Suggestions, the digest's line and cloud walks are charged to the workspace's AI credits; reading bills is not. What the model answers is checked before anything is shown or sent. We keep the result and a record of each call (the model, the tokens it used and its cost), never what was sent.
Feedback, support and error reports
- Feedback you send from the app's Help menu is kept with your account and workspace so we can reply, with the address of the page you were on only if you tick "Include this page's address".
- A support reference you ask for from Help links a short code to your account, your workspace and your last 20 requests' ids, so we can find them in our logs. It expires after 30 days.
- "Was this page helpful?" on the docs records the page, your answer and any comment, and nothing about who you are.
- Error reports: when the app breaks in your browser, it sends us the error's name, its message with values such as addresses, ids and quoted text removed, the names of the functions involved, the kind of page (such as
/app/:slug/projects/:id) and the app's version, and nothing about you. They go to our logs.
Google user data
ShipVitals offers two kinds of Google access:
- Signing in with Google shares your name, email address and profile picture with ShipVitals. We use them to create and identify your account.
- Connecting Google Analytics or Search Console (arriving during the beta), if a workspace admin chooses to, asks for read-only access to Google Analytics reports and Search Console performance data (
analytics.readonlyandwebmasters.readonly). ShipVitals uses it to read the properties and sites you choose and show figures such as active users and search clicks for your projects. The refresh token Google issues is stored encrypted with the workspace's key, every use is recorded in the workspace's audit log, and disconnecting removes it.
ShipVitals' use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements. In particular, we use Google user data only to provide and improve the features you see in ShipVitals; we don't transfer it to anyone except to provide those features, for security or to comply with the law; we don't sell it, don't use it for advertising, don't use it to train AI models, and don't let people read it unless you ask us to, it is needed for security or to comply with the law, or it has been aggregated and anonymised.
Who else processes data
We use a small number of service providers to run ShipVitals, listed with what they do on the Sub-processors page. We don't sell personal data. We disclose it only to those providers, when the law requires it, or to a successor if ShipVitals changes hands, under this policy.
Where data is kept
ShipVitals' database and files are kept in Google Cloud's europe-west1 region, in Belgium. Sign-in accounts (Firebase Authentication) and the service's logs are kept on Google's infrastructure, which may be outside the European Economic Area. Some providers on the sub-processors list process data outside the European Economic Area; where they do, we rely on the transfer safeguards they offer.
How long we keep it
- Your account, until you delete it. Deletion removes your sign-in, memberships and tokens at once and keeps only the account's id with a few dates and settings, with no name or email address, so the account can't be revived.
- A deleted workspace can be restored for 7 days, then its data is deleted and its encryption key destroyed.
- Workspace figures, for the history window of the workspace's plan.
- Audit logs and your account's security records, for 13 months.
- Export files, for 2 days; their download links work for 24 hours.
- Forwarded bills' files, cloud browser screenshots and cloud walks' evidence, for the plan's evidence period (7, 30 or 90 days); a mail marked "Not an invoice", 7 days.
- Support references, for 30 days.
- Database backups roll over within 14 days.
Your choices and rights
You can see and change your profile and preferences in your account settings, download your data with Export data, and delete your account at any time. Depending on where you live, you may also have the right to access, correct, delete, restrict or object to our use of your data, and to complain to your data protection authority. Write to support@nannertech.com and we'll reply.
Cookies and storage
ShipVitals sets no cookies. Your browser keeps your theme and a few interface choices, and while you sign in your email address and any invitation link, in local and session storage, and the sign-in session in storage that Firebase Authentication manages. Our analytics use none of it. Stripe's checkout and billing pages are Stripe's own, under its own cookie policy. See Cookies.
Children
ShipVitals is not meant for anyone under 16, and we don't knowingly collect their data.
Changes
We date every version of this policy and tell you in the app about material changes.