Data Source Connectors
Ragty can synchronize documents from 30 external source types. Connector credentials are encrypted at rest and each data source remains bound to its tenant, workspace, and dataset authorization context.
Synchronization lifecycle
Open Data Sources, choose a connector, and configure its credentials or OAuth authorization.
Enable autosync to queue the first synchronization immediately.
The worker checks for due sources every minute. Recurring synchronization defaults to every 60 minutes and can be configured from 1 minute to 7 days.
Incremental connectors add or update changed content and remove content deleted at the source.
Synchronization uses durable claims so multiple workers do not process the same source concurrently. Failures use bounded retry backoff. Autosync can be paused while Sync Now remains available for an explicit run.
Connector reference
Cloud storage
Connector | Typical authorization | Content |
|---|---|---|
Google Drive | Google OAuth2 | My Drive and Shared Drive files and folders |
OneDrive | Microsoft credentials | OneDrive files |
SharePoint | Microsoft credentials | SharePoint site documents |
Box | Box OAuth2 | Box files and folders |
Dropbox | Dropbox OAuth2 | Dropbox files and folders |
Azure Blob | Connection credentials | Azure Blob containers |
Seafile | Server URL and token | Seafile libraries |
Blob Storage | Access credentials | S3, R2, GCS, OCI, and S3-compatible object storage |
Collaboration and knowledge bases
Connector | Typical authorization | Content |
|---|---|---|
Confluence | Atlassian OAuth2 | Pages, posts, and attachments |
Notion | Integration token | Pages and databases |
Slack | Slack OAuth2 | Channel messages and threads |
Microsoft Teams | Microsoft credentials | Team channel messages |
Asana | Access token | Tasks and comments |
Zendesk | Subdomain and API token | Help Center and support content |
Salesforce | Instance credentials | Cases, knowledge, and configured objects |
DingTalk AI Table | API credentials | AI Table records |
Moodle | Site URL and token | Course and forum content |
Discord | Bot credentials | Server channels |
Development tools
Connector | Typical authorization | Content |
|---|---|---|
GitHub | GitHub OAuth2 | Repository content, issues, and pull requests |
GitLab | Personal access token | Repository content, issues, and merge requests |
Jira | Atlassian OAuth2 | Issues, epics, and comments |
Bitbucket | Username and app password | Repository content and pull requests |
Connector | Typical authorization | Content |
|---|---|---|
Gmail | Google OAuth2 | Messages and attachments |
Outlook | Microsoft credentials | Outlook or Exchange messages and attachments |
IMAP | Host and account credentials | Mail from an IMAP-compatible server |
Databases
Connector | Typical authorization | Content |
|---|---|---|
Airtable | Personal access token | Base and table records |
RDBMS | Database credentials | Selected PostgreSQL or MySQL rows |
Protocols
Connector | Typical authorization | Content |
|---|---|---|
RSS | None for public feeds | RSS and Atom articles |
REST API | Optional bearer token, API key, or basic authentication | JSON API responses |
WebDAV | Server and account credentials | WebDAV files |
OAuth2 setup
OAuth-enabled connectors require an application registration at the provider. Ragty reads the following optional environment-variable pairs:
Register this redirect URI for each provider:
M8TY_RAGTY_FRONTEND_URL must be the browser-visible Ragty frontend URL, including any configured base path.
Origin
Connector implementations are adapted from the RAGFlow project under the Apache License 2.0. Ragty adds its authorization boundary, encrypted credential storage, scheduling, and lifecycle controls.