Privacy Policy

About RepoSentry

RepoSentry is a research repository application operated by the RepoSentry project team. It uses document classification, access controls, and activity records to help invited users manage research data. This policy describes data processed through reposentry.org.

Information we process

Google sign-in provides your name, email address, and profile information. We process account roles, invitations, authentication sessions, authenticator enrollment information, and Google authorization tokens.

When you connect Google Drive, we access file names, identifiers, types, sizes, timestamps, ownership and sharing metadata, and file contents needed for classification or requested downloads. We store repository metadata, classification results and document embeddings. Files uploaded directly to RepoSentry are stored by the service.

Security and activity records can include your account, actions, affected files, IP address, browser information, timestamps, and access decisions.

How we use Google Drive access

Ordinary members use Google for sign-in without granting Drive access. The repository owner connects Google Drive with per-file access (drive.file), which permits creating and managing files created by or explicitly opened with RepoSentry. Repository documents are stored in a dedicated private folder. Scans, classification, monitoring and downloads are limited to that repository; personal Drive files outside it are excluded. The owner retains access in Google Drive. Background access uses the owner’s stored refresh token until it expires or is revoked.

Document processing supports the repository features you use. Google user data is not used to train generalized AI or machine-learning models. RepoSentry does not sell Google user data or use it for advertising.

RepoSentry's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.

Usage analytics and session replay

We use PostHog US Cloud to understand how signed-in users navigate RepoSentry and to investigate usability problems. Analytics use your internal RepoSentry account ID, role, browser/device information, page visits and build version. During an evaluation, the participant code and trial identifier link this activity to the evaluation record.

Session replay starts on protected pages after authenticator verification. It records page layout, interactions and visible page text, which may include names, email addresses, research titles and details displayed on screen. Form values and explicitly private regions remain masked. Authentication screens, images, embedded documents and file inputs are excluded. We disable console recording and request/response body and header capture. Names, email addresses and document contents are not added as analytics properties.

The browser uses session storage for analytics identity and local storage to stop recording when an account changes in another tab. Recordings and analytics are processed in the United States and are available to authorized project team members. Contact the project team about access, deletion or retention of these records. Replay supports observation; questionnaire opinions and formal evaluation results are recorded separately.

Access and service providers

Authorized repository users and administrators can access information according to their roles and repository permissions. The service uses Google for authentication and Drive, Vercel for the website, and Fly.io for backend processing and database hosting. Configured email providers process invitation and notification details when those features are used. Providers process information to operate these services; data may be processed outside your country.

Project administrators may access records when needed to support users, investigate security issues, or administer the repository. We may disclose information when legally required.

Storage and security

We use HTTPS, access controls, and encrypted storage for persisted Google refresh tokens and authenticator secrets. Session cookies support sign-in. These protections reduce risk but cannot guarantee that every security incident will be prevented.

Repository, account, and audit records remain stored until removed by an administrator or through a deletion request. Revoking Google access stops future authorized access but does not automatically delete data already stored in RepoSentry. Backup copies may remain until they are rotated out, and records needed for security or legal obligations may be retained.

Your choices and deletion requests

You can revoke RepoSentry's Google access through Google Account connections. Contact runicoleee@gmail.com to request access, correction, or deletion of your RepoSentry data, or to ask about retention. We may need to verify your identity and coordinate with your repository administrator. Removing a RepoSentry record does not delete the original file in Google Drive.

Policy updates

We will update this page and its effective date when these practices change. Contact the RepoSentry project team at runicoleee@gmail.com with privacy questions.