Confluence Connector
Connect Confluence as a context source for the AI Documentation Agent to search pages with CQL, read page content, and keep your docs aligned with your team's wiki.
Overview
Use the Confluence integration to connect your Atlassian account and make Confluence available as an AI agent context source. The Documentation AI agents read Confluence pages live at runtime. When a prompt needs context from confluence, the agent queries Confluence on demand with CQL, then reads the relevant pages directly.
Before you start
You need:
- The Admin or Editor role in your Documentation.AI organization
- An Atlassian account with access to the Confluence content you want the agent to use
- At least one documentation project created in your Documentation.AI workspace
Connect Confluence
Open Agent Settings
In the dashboard sidebar, under AI, click Agent Settings. The Connectors section lists the tools the agent can pull context from.
If you are in the editor, you can click Connectors below the chat input instead. It opens the same list.
Click Connect on the Confluence card
Atlassian opens in the same tab. Sign in if Atlassian asks you to.
Complete the Atlassian OAuth flow
Choose the Confluence site and approve access. Atlassian sends you back to the page you started from, and "Connected successfully" appears.
Check that the connector is on
The Confluence card now shows your Confluence site name, Manage, and a switch that is on. Confluence is now on for the documentation project you have open, so the Writer and Answer agents can query it.
The Confluence connection belongs to your organization, but the switch is set per documentation project. To use Confluence in another project, open that project, go to Agent Settings, and turn on the Confluence switch.
How the AI agent uses Confluence
The Confluence subagent uses these tools to find and read the content your prompt needs.
| Tool | What it does |
|---|---|
listSpaces | Lists all Confluence spaces accessible to the connected account |
searchPages | Searches pages using CQL by text, space, label, or other criteria |
getPage | Fetches full content of a single page by ID |
listSpacePages | Lists pages within a specific space |
getSpaceTree | Retrieves the page tree hierarchy for a space |
getPageChildren | Lists child pages under a specific page |
getPageLabels | Lists labels attached to a page |
The agent decides which tools to call based on your prompt. For example, it may search with CQL first, inspect a space tree to narrow the scope, then fetch one or more pages to answer a question or write documentation.
Example prompts
Use prompts like these when you want the AI agent to pull context from Confluence:
- Summarize the deployment process documented in our Platform Engineering space.
- Find Confluence pages about SSO setup and update the authentication guide to match them.
- Search for pages labeled onboarding and draft a getting-started checklist for new support agents.
- Compare the incident response steps in Confluence with the current runbook in this documentation project.
Troubleshooting
Use these checks if the Confluence integration or project connector is not behaving as expected.