Privacy Policy

Who is responsible for your data

Classic Source is operated by DattaDock LLC, Oklahoma City, Oklahoma, United States. For any privacy question or request, email info@classicsource.app.

What we collect

If you use photo part lookup

The photo is resized on your device before anything leaves it, and that resize strips the photo's hidden metadata, including any location. The full-size photo goes to Anthropic once, to identify the part, and is not stored by Classic Source. What we keep with your account is a small thumbnail plus what was read off the part, so you can revisit the lookup. Deleting a lookup removes its thumbnail; deleting your account removes them all.

If you add a photo of your vehicle

A photo you add to a car in your garage is stored with your account, against that vehicle, so it appears whenever you open the car. It is kept until you replace it, delete the vehicle, or delete your account, and it is included in the copy of your data you can request. Unlike a part lookup, this one is stored rather than read once and discarded.

If vehicle rendering is available and you explicitly allow it, the first picture is sent to Google's image model to produce a cleaned-up version for the top of the vehicle's page. We also send the vehicle’s year, make and model, but do not attach your name, email or account identifier. The photo itself may show identifying details. Your original photo is always kept and is what the app falls back to, and the vehicle photo menu lets you switch between “Show the photo” and “Show the rendering.” The illustrated version can appear as the cover without a separate visible label.

If you create an account

• Your email address.
• Your name, only if you provide one. It is optional and it is only used to greet you in the app.
• Your password, stored only as a scrypt hash with a random salt. We cannot read your password, and it is never stored in plain text.
• If you sign in with Google, the account identifier Google gives us plus the email address and name on your Google account. We never receive your Google password.
• If you sign in with Apple, the account identifier Apple gives us plus an email address and, the first time only, your name. We never receive your Apple password. If you choose Hide My Email, the address we store is Apple's forwarding address, not your real one.
• Which plan you are on, and the date you created the account.

What you save in the app

• Your garage: the vehicles you save, meaning year, make, model, submodel, engine and an optional note.
• Your saved searches: the title you give them, the vehicle they are scoped to, and the part term.

These are stored against your account and are private to you. If you are not signed in, they are kept in your own browser's local storage instead and never reach our servers.

If you subscribe to a paid plan

• A Stripe customer reference and subscription reference, plus the status of the subscription, so we know which plan your account is on.
• A count of how many AI questions you have used in the current calendar month, so the monthly allowance can be enforced.

Search history, linked to your account

Separately from the counts below, we keep a short rolling log of individual searches so we can help when something goes wrong. Each entry holds the part term you typed, the vehicle you selected, how many results came back, and the email address on your account.

• Only the most recent 300 searches across all customers are kept, and every entry is deleted automatically after 30 days.
• Nothing is logged for signed-out visitors, and no IP address is stored.
• It is visible only on our internal dashboard, to staff. It exists so that "somebody searched a 1965 Corvette and got nothing back" is a problem we can actually follow up on.

Search activity, kept in aggregate

We keep a running count, one record per day, of what people search for: part terms, vehicles, searches that returned nothing, which retailers responded and how many results they returned, and how long searches took. This is how we find out that a retailer has gone quiet or that a common part gets no results, which is what tells us what to fix next.

• These counts are not linked to your account. Your user id is not written into them, and there is no way to read back what any one person searched for.
• To stop a single source from flooding the lists and to avoid counting the same search several times while it is still running, the requesting IP address is folded into a short one-way digest that lives only inside that day's record. The IP address itself is not stored, and the digest is never displayed anywhere.
• Search terms that look like an email address, a phone number or a long account number are redacted or dropped before anything is written, so free text you typed cannot end up in these counts.

What we do not collect

Who else is involved

These services receive data for the purposes described below. External links take you to sites governed by their own privacy policies.

• Netlify hosts the site and runs the application. Standard server logs include request information such as IP addresses. Netlify Blobs holds sessions, caches, rate and spend counters, the search history and diagnostic events described here, and a standby copy of account records and saved app data so the service can work if the database is unreachable.
• Supabase hosts our database, including your account, garage, saved searches and parts, photo lookups, saved conversations, vehicle photos, notes and build lists.
• Resend delivers account verification, password-reset, welcome, team-invitation and requested reminder emails. It receives the recipient address, any name or vehicle/reminder details in the message, and the message contents, including verification and reset links. New-account notifications sent to the operator also contain the new account’s email address and optional name.
• Stripe processes subscriptions bought on our website. It receives your email address and payment information; card details are entered directly on Stripe’s page. Stripe keeps payment records under its own privacy policy.
• Anthropic receives data only after you allow Anthropic AI sharing in the app. Depending on the feature, this includes your question or search term, selected vehicle, product listings, conversation history, garage, saved parts, build list, recent searches, vehicle notes and attached photos. This powers answers, part and VIN-tag identification, job plans, install guides and search assistance. We do not attach your account ID, name or email, but images and text you supply may themselves identify you. Anthropic processes requests under its API terms, including its retention and abuse-monitoring rules. We do not use these requests to train an AI model. Conversations you choose to save remain in your account.
• Google handles Google sign-in on the website if you choose it; its sign-in script can load when the sign-in dialog opens. Google sends us your account identifier, email and name. Google sign-in is not offered in the iOS app.
• Apple handles Sign in with Apple if you choose it and sends a signed account identifier and email, plus your name when available. Hide My Email keeps your real email private from us. On account deletion we request revocation of Apple access. If Apple is temporarily unavailable, revocation is retried; if no usable token exists, the app explains how to remove access in Apple account settings. The app reports the outcome without keeping your Classic Source account open.
• Google’s Gemini image service receives the first vehicle photo plus its year, make and model only after you explicitly allow Google rendering. We do not attach your account ID, name or email; the image can contain a license plate or other identifying details. Declining saves your original photo without sending it to Google. The photo menu lets you compare the original with an illustrated rendering; a cover rendering does not have a separate visible label.
• Zyte fetches some retailer catalogs on our behalf. It receives retailer URLs containing the search terms and vehicle details needed for that catalog lookup. We do not send your account ID, email or device IP with those requests.
• NHTSA receives the VIN you ask us to decode through its public vPIC lookup. The request comes from our server without your account details or device IP. Photographing a VIN can also use Anthropic to read the image, after permission.
• YouTube (Google) receives vehicle and part terms from our server to find install videos. Thumbnails are proxied by us. Playing a video contacts YouTube’s embedded player, and following a watch link opens YouTube; Google then receives the normal browser request information under its own policy.
• Retailers receive catalog queries from our servers; product images are proxied through our server. Clicking a store link opens its website, which receives your browser request under its own policy. Some links earn us a commission.
• Impact handles affiliate redirects for Summit Racing links. Clicking one sends Impact the destination URL and normal browser request information before it forwards you to the retailer, so a qualifying purchase can be attributed to our referral. Amazon affiliate links also carry our referral tag directly to Amazon. These tags do not affect result ranking.
• Hagerty, Barn Finds and Silodrome supply news headlines through generic feed requests from our servers. They receive your browser request only if you open an article link.

Your AI choices

Before sending data to Anthropic or Google’s image service, we name the provider and explain the data and purpose in a permission sheet. Permission is kept for your account in this browser session and is cleared when you sign out or change accounts. Use Reset AI permissions in Account to withdraw it for future requests. This does not recall a request already sent. Ordinary search, saved records and storing an original vehicle photo remain available when you decline.

Service diagnostics

We keep up to 400 recent events per measured feature to understand failures and performance. An event can include its time, outcome, duration, error, plan, photo size and limited feature metadata. A stable digest of your user ID links repeated events from the same account; because we can compute that digest from account records, we treat these diagnostics as linked to you. They are used to operate and improve the service, not for advertising or tracking across other apps.

We may also disclose information if we are legally required to, or where it is necessary to investigate abuse or protect the service.

Cookies and sessions

• Signing in sets one cookie, named cs_session. It holds a random token and nothing else. It is HttpOnly, Secure, SameSite=Lax, and it expires after 30 days.
• The session token is stored on our side only as a SHA-256 hash, so a copy of our storage would not yield a usable session.
• Signing out invalidates the session.
• There are no advertising or tracking cookies. We do not use a cookie banner because there is nothing optional to consent to.
• If you are signed out, your garage and saved searches are held in your browser's local storage. Clearing your browser data clears them.

How long we keep things

• Account data, garage and saved searches: for as long as your account exists, until you ask us to delete it.
• Sessions: 30 days, then they expire.
• Aggregate daily search counts: 30 days, after which the day's record is automatically swept.
• Billing records: Stripe keeps payment records under its own retention rules, and we may keep basic records of a subscription for as long as tax and accounting rules require.

Getting a copy of your data, or having it deleted

You can delete your account yourself, at any time. Sign in, then use Delete account in the account menu. It removes your account record, your garage, your saved searches, your saved parts and your usage counts straight away, and cancels any subscription. It cannot be undone.

If you would rather we did it, or you want a copy of your data first, email info@classicsource.app from the address on your account and tell us what you want. We aim to handle requests within 30 days.

Two honest caveats. The aggregate daily counts cannot be traced back to an individual, so there is nothing in them to pull out, and they age out on their own within 30 days. And if you have subscribed, Stripe holds its own payment records under its own policy, so you may need to contact Stripe as well for those.

Depending on where you live you may have additional rights over your data, such as access, correction, deletion or objecting to processing. Write to us at the address above and we will handle the request the same way.

Security

The site is served over HTTPS. Passwords are hashed with scrypt, session tokens are stored hashed, API keys are held server side only and never reach your browser, and the site sets a strict content security policy. No system is perfectly secure, but we are not holding payment data, and that is deliberate.

Children

Classic Source is not directed to children and is not intended for anyone under 13. We do not knowingly collect information from children. If you believe a child has created an account, email us and we will remove it.

Where data is handled

DattaDock LLC is based in the United States, and the services above process data in the United States and other countries where their infrastructure operates. If you use Classic Source from outside the United States, your information will be processed in the United States.

Changes to this policy

If we change what we collect or who we share it with, we will update this page and change the "last updated" date at the top. If a change is significant and you hold an account, we will tell you by email.