Gleap AI agents can work in Confluence through Composio, for product and engineering teams that keep runbooks, specs and known issues in Confluence while customer reports and feature requests arrive in Gleap. A custom agent can search those pages and add customer evidence to them, and Kai Code can read your internal docs while it works on a fix.
What a Gleap agent can do in Confluence
- Runbooks while a ticket is open: CQL Search finds the runbook or known-issue page that matches a Gleap bug report, so the teammate sees the documented workaround.
- Customer evidence on specs: Create Footer Comment adds a comment to a spec page with what customers asked for in Gleap feature requests, in their own words.
- Known-issue pages: Update Page can refresh an internal known-issues page after a teammate confirms a new bug.
- Docs for coding sessions: in Kai Code, Get Page by ID and Search Content give the coding agent your architecture and API pages as context.
- Starting on page changes: Confluence triggers such as New Page Created and Page Updated can start a custom agent when documentation changes.
Example: a workaround that already exists
A customer reports through the Gleap widget that CSV imports fail for files with semicolons. A teammate asks the agent on the ticket to check the docs. The agent runs a CQL search, finds a known-issue page with a documented workaround and posts the link and the steps in the ticket’s agent chat. The teammate decides whether to share the workaround with the customer.
How to connect Confluence
- In Gleap, open AI & Automation, then Agents, and select the custom agent that should use Confluence, or click Create agent.
- Under Tools, click Add tool, search for “Confluence” and select it.
- Click Connect Confluence and follow the authorization Gleap shows: approve access in the Atlassian sign-in window, or enter an API token if the form asks for one.
- Open the Confluence tool and switch off every action the agent should not use, such as Update Page, Delete Space and the space and property changes.
- Publish the agent and test it in chat before you turn on a trigger.
For Kai Code, open Settings, then MCP, click Add MCP server and search for Confluence.
Kai Code is on Team, Pro and Enterprise, and Kai Support can use connected apps as tools on the same plans. Custom agents are on the Team, Pro and Enterprise plans.
FAQ
Does Gleap integrate with Confluence?
Yes. Gleap AI agents and Kai Code use Confluence through Composio’s Confluence toolkit, which supports Atlassian OAuth and API tokens. The agent reads or writes pages while it runs; Confluence content is not imported into Gleap’s knowledge base.
Does Gleap connect to Confluence through MCP?
Yes. Gleap connects to Composio’s MCP server for Confluence with the Confluence account you connect, and the agent calls only the actions you leave switched on. To use a different remote MCP server, add it by URL: as a Custom MCP tool for Kai Support or a custom agent, or as a custom HTTP server in Kai Code.
What can a Gleap AI agent do in Confluence?
It can search with CQL, read spaces, pages, versions, labels and comments, create and update pages and comments, add labels and manage spaces and properties. Update Page replaces the whole page body, so give that action only to agents that should edit documentation.
Which Gleap plan includes the Confluence integration?
Kai Code can use Confluence on Team, Pro and Enterprise, and Kai Support can use connected apps as tools in its customer conversations on the same plans. Custom agents can use it on Team, Pro and Enterprise. Confluence space permissions of the connected user still apply. AI usage is billed separately; see pricing.
Should customer-facing answers come from Confluence?
Usually not. Internal pages can contain details customers should not see. Keep customer answers in Gleap’s knowledge base, which Kai Support answers from, and use Confluence for internal context that a teammate reviews first.
