Confluence Connector

The Confluence connector ingests wiki pages, blog posts, and their updates into CortexDB as document episodes. This gives your AI systems access to your organization's knowledge base.

Setup

CORTEX_CONFLUENCE_URL=https://your-org.atlassian.net/wiki
[email protected]
CORTEX_CONFLUENCE_API_TOKEN=your-confluence-api-token
CORTEX_CONFLUENCE_SPACES=ENG,PRODUCT
CORTEX_CONFLUENCE_TENANT_ID=my-app

cortexdb-connector confluence \
  --url https://your-org.atlassian.net/wiki \
  --email [email protected] \
  --token your-confluence-api-token \
  --spaces "ENG,PRODUCT" \
  --tenant-id my-app

What Gets Ingested

| Confluence Event | Episode Type | Content | |---|---|---| | Page created | document | Page title and body (converted from storage format to plain text) | | Page updated | document | Updated content with change metadata | | Blog post | document | Post content | | Comment | message | Comment text |

Episode Example

{
  "type": "document",
  "content": "# API Rate Limiting Policy\n\nAll API endpoints are rate-limited per tenant. Default limits:\n- 1000 requests/second for standard tier\n- 5000 requests/second for enterprise tier\n\nRate limits are enforced at the API gateway...",
  "source": "confluence",
  "author": "bob",
  "timestamp": "2026-03-10T14:00:00Z",
  "metadata": {
    "page_id": "12345",
    "space_key": "ENG",
    "title": "API Rate Limiting Policy",
    "version": 3,
    "last_edited_by": "bob",
    "labels": ["api", "policy", "rate-limiting"],
    "url": "https://your-org.atlassian.net/wiki/spaces/ENG/pages/12345"
  }
}

Configuration

| Variable | Default | Description | |---|---|---| | CORTEX_CONFLUENCE_URL | Required | Confluence base URL | | CORTEX_CONFLUENCE_EMAIL | Required | User email | | CORTEX_CONFLUENCE_API_TOKEN | Required | API token | | CORTEX_CONFLUENCE_SPACES | Required | Comma-separated space keys | | CORTEX_CONFLUENCE_TENANT_ID | Required | Target tenant | | CORTEX_CONFLUENCE_NAMESPACE | confluence | Target namespace | | CORTEX_CONFLUENCE_INCLUDE_ARCHIVED | false | Include archived pages | | CORTEX_CONFLUENCE_INCLUDE_COMMENTS | true | Include page comments | | CORTEX_CONFLUENCE_BACKFILL_DAYS | 365 | Days of history to backfill | | CORTEX_CONFLUENCE_POLL_INTERVAL_SEC | 300 | Polling interval (5 min default) |

Content Processing

Confluence stores page content in a proprietary storage format (XHTML). The connector automatically:

1. Converts storage format to clean markdown
2. Extracts text from tables and macros
3. Preserves headings and structure
4. Strips Confluence-specific markup