Confluence

Use the Confluence Connector to ingest pages from Atlassian Confluence Cloud or Server/Data Center into the mAItion knowledge base.

What It Does

discovers pages via a space, a list of page IDs, a label, a CQL query, or a folder
converts each page to Markdown for indexing
optionally follows descendant pages when discovering by page IDs
runs ingestion on configurable schedules

Discovery Modes

Exactly one discovery mode must be set per source:

Mode	Config key	Description
Space	`space_key`	All pages in a space
Page IDs	`page_ids`	Comma-separated list of page IDs
Label	`page_label`	All pages with a given label
CQL	`cql`	Arbitrary CQL query
Folder	`folder_id`	All pages inside a folder

Authentication

Set one of the following. Username/password, API token, and bearer token auth are mutually exclusive with oauth2; cookies is the exception and may be combined with oauth2.

Cloud basic (username + api_token): email address and API token. Recommended for Confluence Cloud. Generate a token at id.atlassian.com/manage-profile/security/api-tokens.
Server basic (username + password): account username and password. For Confluence Server/Data Center.
Bearer token (api_token only): a Personal Access Token. For Server/Data Center; requires cloud: false.
OAuth2 (oauth2): an OAuth2 credentials dict (client_id, token) passed directly to the underlying reader.
Cookies (cookies): a session cookies dict, e.g. produced by atlassian.utils.parse_cookie_file(). Can be combined with oauth2.

Required Environment Variables

Set these in .env.rag:

CONFLUENCE1_BASE_URL: Confluence base URL (example: https://yoursite.atlassian.net/wiki)
CONFLUENCE1_USERNAME: Confluence username or email — required for basic auth
CONFLUENCE1_API_TOKEN: API token (Cloud basic auth) or Personal Access Token (Bearer auth) — mutually exclusive with CONFLUENCE1_PASSWORD
CONFLUENCE1_PASSWORD: Confluence password (Server basic auth) — mutually exclusive with CONFLUENCE1_API_TOKEN
CONFLUENCE1_SPACE_KEY: space key to ingest — set when using the space discovery mode
CONFLUENCE1_PAGE_IDS: comma-separated page IDs — set when using the page IDs discovery mode
CONFLUENCE1_PAGE_LABEL: page label — set when using the label discovery mode
CONFLUENCE1_CQL: CQL query — set when using the CQL discovery mode
CONFLUENCE1_FOLDER_ID: folder ID — set when using the folder discovery mode
CONFLUENCE1_SCHEDULES: ingestion interval in seconds (default is 3600)

`config.yaml` Example

sources:
  - type: "confluence"
    name: "confluence1"
    enabled: true  # optional, default: true
    config:
      base_url: "${CONFLUENCE1_BASE_URL}"
      username: "${CONFLUENCE1_USERNAME}"
      api_token: "${CONFLUENCE1_API_TOKEN}"
      space_key: "${CONFLUENCE1_SPACE_KEY}"
      page_status: "current"        # optional: filter by status, space_key mode only
      max_pages: 50                 # optional, default 50
      schedules: "${CONFLUENCE1_SCHEDULES}"
      request_delay: 0.1            # optional, delay in seconds between requests (default: 0)

# .env.rag

CONFLUENCE1_BASE_URL=https://yoursite.atlassian.net/wiki
CONFLUENCE1_USERNAME=you@example.com
CONFLUENCE1_API_TOKEN=your-api-token
CONFLUENCE1_SPACE_KEY=ENG
CONFLUENCE1_SCHEDULES=3600

Configuration Reference

Field	Required	Default	Description
`enabled`	no	`true`	Set to `false` to skip this source entirely
`base_url`	yes	—	Confluence base URL
`api_token`	see Authentication	—	API token (Cloud basic) or Personal Access Token (Bearer); mutually exclusive with `password`
`username`	see Authentication	—	Confluence username or email
`password`	see Authentication	—	Confluence password (Server basic); mutually exclusive with `api_token`
`cloud`	no	`true`	`true` for Confluence Cloud, `false` for Server/Data Center
`oauth2`	see Authentication	—	OAuth2 credentials dict; mutually exclusive with token/password auth
`cookies`	see Authentication	—	Session cookies dict; can combine with `oauth2`
`space_key`	one discovery mode required	—	Load all pages from a space
`page_ids`	one discovery mode required	—	Comma-separated page IDs to load
`page_label`	one discovery mode required	—	Load all pages with this label
`cql`	one discovery mode required	—	CQL query to select pages
`folder_id`	one discovery mode required	—	Load all pages inside this folder
`page_status`	no	—	Filter by page status, e.g. `current` (space_key mode only)
`include_children`	no	`false`	Also load descendant pages (page_ids mode only)
`max_pages`	no	`50`	Maximum pages to load
`schedules`	no	`3600`	Ingestion interval in seconds
`request_delay`	no	`0`	Seconds to wait between API requests. Increase to avoid rate-limiting (e.g. `0.1`)

Multiple Confluence Sources

Add more sources entries (confluence2, confluence3, etc) with separate env vars per source.

Getting started

Features

Connectors

Tools

Configuration

What It Does

Discovery Modes

Authentication

Required Environment Variables

`config.yaml` Example

Configuration Reference

Multiple Confluence Sources

​What It Does

​Discovery Modes

​Authentication

​Required Environment Variables

​config.yaml Example

​Configuration Reference

​Multiple Confluence Sources

What It Does

Discovery Modes

Authentication

Required Environment Variables

`config.yaml` Example

Configuration Reference

Multiple Confluence Sources