Privacy

Your account and work

Account and session information is used to sign you in and protect access. The application stores book content, images, project settings, access permissions, publication history and reading progress when you use those features.

Drafts require authorised access, including any people to whom you grant read or write permission. When a book is approved for publication, its published content and metadata become available to readers. Reviewers can access material submitted for moderation.

To investigate reported problems, the superadministrator can use a time-limited support session to view and change an author’s application content. The service records who acted, for which account and what kind of change was made. This does not give access to the author’s password or sign-in settings.

Services and integrations

Cloudflare hosts the application and its data. Clerk provides account authentication. If you choose Google sign-in, basic profile information such as your name and email address is used to identify and manage your account. Connecting an external assistant through MCP gives that client the access you approve; its own handling of information is governed by its provider. Connection records include the client and its approved permissions. Revoking access stops future authorised requests, but does not delete information already received by the client.

Service backups contain application data and are stored in private Cloudflare storage. Only the service’s super administrators can manage these copies and use them for recovery.

Use of Google data

Writers Graph uses Google account information for sign-in and account management. It does not sell Google user data or use it for advertising or training AI models. Information is handled for the requested features, their security and applicable legal requirements. A connection to an external MCP assistant is separate and remains subject to that provider’s data practices.

Reading statistics

Optional reading statistics are separate from the progress used to resume reading. If enabled, they use a pseudonymous identifier derived from your account ID; this is not full anonymisation. Turning them off removes your individual statistics progress, while aggregate totals remain. Detailed statistics have a 90-day retention policy with periodic cleanup; physical deletion may be delayed.

Storage and technical information

The application uses browser storage for preferences, local reading state and unsaved document drafts. Local drafts can contain text; they are not just settings. You can remove local drafts in your account settings. Local drafts older than 30 days are cleaned up when you use the application. Authentication cookies maintain sign-in. Service providers process technical request information to deliver and protect the service. Operational measurements help monitor errors, usage and resource limits. Browser telemetry reports categories of events rather than document content and respects browser Do Not Track and Global Privacy Control signals. IP addresses may be processed for request limiting and by infrastructure providers.

These information pages do not load sign-in code, analytics scripts or application APIs.

Retention and requests

Information is kept for the features you use and for security, moderation and operational records. Deleting content does not necessarily remove it immediately from existing published editions, caches or private backups. Deleting a sign-in account does not automatically erase every application record. Reading progress used to resume a book is separate from optional statistics and has no automatic time-based expiry.

Contact contact@writersgraph.com to ask about your information or request access, correction or deletion. Requests are assessed for the relevant account and data; identity verification may be needed. Existing backups follow their configured rotation; withdrawing a publication does not recall copies already downloaded by readers.