What Hutch Is
Hutch (also called HutchDB) is a structured data storage service for AI agents and humans. The hosted version at hutchdb.com and app.hutchdb.com is operated by Expedited Security Inc. This policy explains what data we collect, why, who receives it, how long we keep it, and what you can do about it. It also covers Hutch when you use it through an AI assistant, such as the HutchDB plugin in ChatGPT.
Data We Collect
Account information: When you sign up, we collect your name and email address. If you sign in with OAuth (GitHub, Google), we also receive your profile image. We do not store your OAuth provider password. If you sign up with an email and password, we store a one-way hash of the password, never the password itself.
Data you store: Hutch stores the structured data (collections, records, views) that you or your agents create through the API, an AI assistant, or the web interface. This data belongs to you.
Metadata about your data: To run the service, Hutch keeps information about the data you store: record IDs, the time each record was created and last updated, each record's status, collection names, settings, and inferred field types, and whether a collection is published. Some of this, such as record IDs, timestamps, and status, is returned to you and to the AI assistants you connect as part of normal query results.
Sharing information: If you invite people to a collection or organization, we store their email address, the role you gave them, and the state of the invitation.
Connected apps: When you connect an AI assistant or other app to Hutch (for example ChatGPT, Claude, or Codex), we store the app's registered name and redirect URL, the access level you approved (Read-only or Read + write), and the access tokens issued to it.
AI assistant requests: When an AI assistant uses Hutch on your behalf, we receive only the tool requests it sends, such as the records to save or the filters for a query. We do not receive, request, or store your conversation with the assistant.
Usage data: We collect basic server logs (IP addresses, request timestamps, user agents) for security and debugging purposes. Sign-in sessions record the IP address and user agent that started them. We may keep a keyed hash of an IP address to enforce rate limits. When something breaks, an error report with technical details about the failed request is sent to our error-monitoring service. The web app also uses cookieless, aggregate page-view analytics.
What we don't want: Hutch isn't built for payment card data, health information, government ID numbers, or passwords, API keys, and other secrets, and our Terms prohibit storing them. We never ask for them.
How We Use Your Data
- To provide and maintain the Hutch service, including answering the requests your AI assistants send
- To authenticate you and secure your account
- To prevent abuse, for example by rate-limiting requests
- To find and fix errors
- To send emails you trigger, such as collection invitations
- To understand overall usage of the web app, in aggregate
- To respond to support requests
We do not sell your data. We do not use your stored data to train AI models. We do not use it for advertising or build profiles of you from it. We do not share your data with third parties except as required to operate the service (hosting, database, email delivery, error monitoring, and analytics, listed below) or as required by law.
Who Receives Your Data
The hosted Hutch uses the following third-party services to operate:
- Vercel — hosting and deployment
- Neon — PostgreSQL database
- GitHub / Google — OAuth authentication (when used)
- Fastmail — delivery of emails Hutch sends, such as invitations
- Honeybadger — error monitoring
- Plausible — cookieless, aggregate analytics for the web app
Each of these services has its own privacy policy. We only share the minimum data necessary for them to function.
Your data also goes where you send it. AI assistants you connect receive the data their requests return, and their own privacy policies govern what they do with it. People you share a collection with can see and, depending on their role, change it. A collection you publish can be viewed by anyone with its URL.
Data Retention
Your account and stored data persist until you delete them or request deletion. Server logs are retained for up to 30 days. If you delete your account, we will remove your data within 30 days.
A record you delete disappears from queries, views, and exports right away. Its stored copy is erased when you delete its collection or your account. Deleting a collection permanently removes the collection and all of its records.
Connected-app access lasts until you revoke it, or until the app goes a year without using it. An app that registers with Hutch but never finishes connecting is removed after 24 hours.
Your Rights and Controls
You can export your data at any time via the Hutch API, or by asking a connected AI assistant to export a collection as CSV or JSON. You can request deletion of your account and all associated data by emailing us. We will process deletion requests within 30 days.
- Edit or delete records and collections yourself, in the web app or through a connected assistant.
- Choose Read-only access when you connect an app, so it can read your data but not change it.
- Revoke any connected app under Connected apps in the web app. Its access stops immediately.
- Unpublish a collection, or remove people you shared it with, at any time.
Cookies
Hutch uses a session cookie to keep you signed in. We do not use tracking cookies or third-party analytics cookies.
Self-hosted Hutch
This policy covers only the hosted service at hutchdb.com. If you self-host Hutch on your own infrastructure, your privacy practices are your own — Hutch the software does not transmit data outside your server.
Changes
We may update this policy from time to time. We will notify users of significant changes via email or an in-app notice.
Contact
Questions about this policy? Email mike@expeditedsecurity.com.