Skip to content
Leverage AI Developer Support

All Connectors

This page mirrors the shared connector registry at a readable level. Runtime support still depends on API runtime extensions and provider-specific implementation, not just the presence of a registry entry.

Implemented Connectors

Section titled “Implemented Connectors”
Connector Key Default mode Runtime shape
Google Calendar google-calendar live Live calendar event list/search/recent lookup through Google Calendar API; no indexed calendar storage.
Google Contacts google-contacts live Live saved-contact search/list through Google People API; no indexed contact storage.
Confluence confluence live Live Confluence spaces/pages query through Cloud OAuth, Data Center user PATs, or Data Center service/bot PAT access.
GitHub Issues github-issues live Live GitHub issue search/list/detail and comments through the shared work-item domain; no indexed storage.
GitHub Pull Requests github-pull-requests live Live GitHub PR search/list/detail, reviews, review comments, and changed files through work-item.
GitHub Repositories github-repositories live Live GitHub repository, code file, bounded file-read, and recent commit access through code-repository.
Gmail gmail live Live mail query plus optional indexed sync.
Google Drive google-drive live Live file search by default plus optional indexed file sync.
Jira jira live Live work-item query through Atlassian OAuth, Jira Data Center user PATs, or Jira Data Center company-managed service/bot PAT access.
Microsoft Calendar microsoft-calendar live Live calendar event list/search lookup through Microsoft Graph; separate from Outlook mail.
Microsoft Contacts microsoft-contacts live Live Microsoft 365 people search through Graph Search plus saved-contact listing; separate from Outlook mail.
OneDrive onedrive live Live file search by default plus optional indexed Microsoft Graph file sync.
SharePoint sharepoint live Live selected-site pages, document-library files/folders, lists, list items, columns, and bounded file/page reads through Graph.
OneNote onenote live Live selected-notebook section/page inventory, hierarchy, page detail, and bounded page content reads through Graph.
Odoo odoo live Live Odoo JSON-2 business-record query through user-managed or deployment API-key credentials; no indexed storage.
Outlook outlook live Live mail query plus optional indexed sync.
Slack slack live Live-only workspace chat search through Slack RTS; no indexed storage.

Planned Registry Entries

Section titled “Planned Registry Entries”

The planned entries below are seed/catalog placeholders, not implemented runtime paths:

  • Google Chat
  • Salesforce Accounts
  • Salesforce Cases
  • Salesforce Opportunities
  • HubSpot Contacts
  • HubSpot Companies

Use Planned Connectors for the short status and scope notes for each planned entry.

Screenshot Placeholder
Connector catalog with implemented and planned entries

Replace later with a screenshot of the connector catalog once the final catalog grouping and planned-entry display are settled.

How To Read This

Section titled “How To Read This”
  • implementationStatus: "implemented" means the shared domain registry marks the connector as current.
  • Runtime support still comes from apps/api/src/connectors/connectorRuntimeExtensions.ts and provider manifests.
  • Connectors with indexed support retain extracted chunk text and vectors for retrieval when indexed mode is enabled. Live-only connectors do not create indexed copies.
  • User-managed API-key/PAT connectors such as Odoo, Jira Data Center, and Confluence Data Center use API-owned encrypted per-user credential storage. Deployment API-key connectors such as Odoo, Jira Data Center, and Confluence Data Center company-managed access use customer CMS credential profiles and customer-owned Secret storage instead of Better Auth linked accounts.
  • Confluence is live-only on the shared knowledge-page domain. It requires an explicit saved spaces selection, supports bounded spaces/pages/comments/labels/tree/attachment-metadata reads, and does not create indexed page-content storage.
  • SharePoint is live-only across knowledge-page, file-resource, and business-record. It requires selected sites, supports optional selected lists/drives/folders under those sites, and does not create indexed page/list/file storage.
  • OneNote is live-only on knowledge-page. It requires selected notebooks, supports optional selected sections, and does not create indexed notebook/page storage.
  • Planned connectors intentionally have no executable runtime extension and should not appear as addable production connectors unless customer CMS policy and runtime readiness both allow it.

Canonical Sources

Section titled “Canonical Sources”
  • packages/domain/src/connectors/connectorRegistry.ts
  • packages/domain/src/connectors/connectorRegistry.test.ts
  • apps/api/src/connectors/connectorRuntimeExtensions.ts
  • docs/features/connectors/spec.md