This policy explains how DocQuery (“DocQuery”, “we”, “us”), the service at docquery.xyz, collects, uses, stores and shares information when you use it, including the information it receives from your Google account when you connect Gmail. Questions about it can be sent to privacy@docquery.xyz.
In short
- With your permission, DocQuery reads your Gmail to find the documents attached to your email, and files them in a private library that only you can see.
- We use your Gmail data only to provide that library: to show your documents, catalogue them, search them and answer your questions about them.
- We never sell your data, never use it for advertising, never use it to train AI models, and never let people read it without your permission.
- You can delete any document, disconnect a mailbox, or delete your account at any time, and the data goes with it.
Data we receive from Google
DocQuery asks for one Google permission, read-only access to Gmail (the gmail.readonly scope, which Google’s consent screen describes as “View your email messages and settings”). It cannot send, change or delete your email. With that permission, DocQuery accesses:
- Your Gmail address, to tell your connected mailboxes apart.
- Messages that carry a document. DocQuery searches your mailbox (outside spam and trash) for messages with an attachment of a supported type: PDF, Word, Excel, PowerPoint, OpenDocument, RTF, plain text or CSV. For each such message it reads the sender’s name and address, the subject, the date, and Gmail’s message and thread identifiers.
- The attachments themselves, the files of the types above.
Gmail returns the whole of each matching message, including its text, but DocQuery does not store the body of your emails. It does not download images, calendar invitations, archives or other attachment types, and it does not look at messages without a document attached beyond Gmail’s search results.
How we use your Google data
We use the data we receive from Google only to provide the features you see in DocQuery:
- Your library. Each document is stored, so you can open the original, together with the sender, subject and date it arrived with.
- Cataloguing. The text of each document is extracted and an AI model writes its title, summary, category, date and key points. Documents without a text layer, such as scans, are transcribed by an AI model.
- Search. Document text is split into passages and turned into numerical representations (embeddings) so that you can search by meaning as well as by keyword.
- Answers. When you ask a question, the relevant document text is given to an AI model to answer it, with references to the documents used. Your conversations are saved so that you can return to them.
- Keeping your library current. DocQuery checks connected mailboxes for new documents once a day, when you open your library, and when you ask it to.
Summaries, embeddings, extracted text and other information derived from your Google data are treated exactly like the data itself and are used only for these features.
What we never do with your Google data
- We do not sell it, or transfer it to advertising platforms, data brokers or information resellers.
- We do not use it to serve advertisements, including retargeting, personalised or interest-based advertising.
- We do not use it to determine creditworthiness or for lending purposes.
- We do not use it to develop, improve or train generalised or non-personalised AI or machine-learning models, and we route it only to AI providers that have agreed not to train on it.
- We do not use it for any purpose other than providing and improving the DocQuery features described above.
Who can see your Google data
Your library is visible only to you, once you have signed in. Every request is checked against your account, and originals are kept in private storage that is only served to you. No person at DocQuery reads your email, documents, summaries or questions, except:
- where you have asked us to and agreed to let us see specific data, for example to look into a problem with a particular document;
- where it is necessary for security purposes, such as investigating abuse or a security incident;
- where it is necessary to comply with applicable law; or
- where the data has been aggregated and anonymised and is used for internal operations, such as counting how many documents are processed.
How we share your Google data
We share your data only with the service providers that run DocQuery on our behalf, and only so that they can provide the features above. Each is bound by contract to use the data only to provide its service to us:
- Vercel hosts the application, runs its background work, and stores your original documents in private storage. Its AI Gateway passes requests to the AI providers below.
- OpenAI provides the AI models that catalogue documents, transcribe scans, answer your questions and create the embeddings used for search.
- Our database hosting provider stores your library’s records: the details of each document, its extracted text, catalogue entry and embeddings, and your saved conversations.
Requests to AI models are sent only to providers that have agreed with Vercel to keep no copy of the data once the request is complete (“zero data retention”) and not to train on it. If no such provider is available, the request is not made.
Beyond these providers, we disclose your data only where the law requires it, to protect the security of DocQuery and its users, or, as part of a merger, acquisition or sale of assets, after first obtaining your explicit consent.
Google API Services User Data Policy
DocQuery’s 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.
How we store and protect your data
- The permission Google gives DocQuery (a refresh token) is encrypted with AES-256-GCM before it is stored, and is used only to read your mailbox for the features above.
- Original documents are kept in private storage that is not publicly reachable; they are served only through DocQuery, to you, after you have signed in.
- All data travels over encrypted connections (HTTPS/TLS), and our providers encrypt the data they store.
- Access to production systems is limited to the people who operate DocQuery. Application logs identify documents by internal identifiers, not by their names or content.
No method of transmission or storage is completely secure, but we take reasonable technical and organisational measures to protect your data and will tell you without undue delay if a breach affects it.
Retention and deletion
We keep your Google data for as long as the mailbox it came from stays connected, so that your library remains available. You decide when it goes:
- Delete a document and its stored original, text, catalogue entry, embeddings and related conversations are erased. A minimal record (Gmail’s identifiers for the message, its date, and a fingerprint of the file) is kept so that the document is not filed again, until you disconnect the mailbox.
- Disconnect a mailbox on the Sources page and everything that came from it is erased and DocQuery’s access to it is revoked with Google.
- Delete your account and all of your data, from every mailbox, is erased and DocQuery’s access is revoked.
Records are removed from our systems straight away (after an account is deleted, by a background job that starts at once), and stored originals are removed by a background job shortly afterwards. Copies may remain in our providers’ encrypted backups for a limited period until those backups expire.
You can also withdraw DocQuery’s access at any time from your Google Account permissions. DocQuery then stops reading that mailbox; documents already in your library stay there until you delete them or disconnect the mailbox. To ask us to delete your data, write to privacy@docquery.xyz.
Other information we collect
- Account information. Your name, email address and sign-in details, held by our authentication provider, Clerk, when you create an account.
- Your questions and conversations with DocQuery, stored so that you can return to them.
- Usage counts, such as how many questions you have asked today, used to apply the beta’s usage limits.
- Technical information such as IP address, browser type and request logs, collected by our hosting provider to run and secure the service.
We use this information to provide and secure DocQuery, to apply its usage limits, to communicate with you about your account, and to meet legal obligations. We do not sell it.
Cookies
We use only cookies that are needed for DocQuery to work: those that keep you signed in, and a short-lived cookie that protects the Gmail connection process. We do not use advertising or tracking cookies.
Your rights
Depending on where you live, you may have the right to access, correct, export or delete your personal information, and to object to or restrict certain uses of it. You can delete documents, mailboxes and your account yourself in DocQuery, or write to privacy@docquery.xyz and we will respond within 30 days.
International transfers
Our providers may process your information in countries other than the one you live in, including the United States. Where that happens, we rely on appropriate safeguards, such as standard contractual clauses, to protect it.
Children
DocQuery is not directed to children under 16, and we do not knowingly collect personal information from them.
Changes to this policy
When we change this policy we will update the date at the top of this page. If we change how we use data received from Google, we will tell you in DocQuery and ask for your agreement again before using your data in the new way.
Contact
For questions about this policy or your information, or to make a request about your data, write to privacy@docquery.xyz. Your use of DocQuery is also subject to our Terms of Service.