Confluence MCP server: setup, auth, production
Last verified: June 2026· Confluence MCP integration
Community MCP servers existYour organization’s wiki and documentation.
What is the Confluence MCP server?
A Confluence MCP server implements the Model Context Protocol and exposes Confluence's capabilities as typed tools an agent can discover and call in the middle of a task. Your organization’s wiki and documentation.
Docs and comms access give an agent the context that lives outside the repo — the spec, the decision, the channel where it was made — and a way to report back. Scope to the right spaces and prevent noisy or unauthorized posting.
A real Confluence workflow
The agent reads the design doc and ADRs for context before making a change, and generates the documentation page for what it shipped — keeping the wiki current instead of stale.
What an agent can do with Confluence
- Have an agent read design docs for context
- Generate documentation pages from a change
- Search internal knowledge while working
Read-only is the right default: let the agent observe and reason first, then grant specific write actions deliberately — each behind audit logging and, for anything high-impact, human approval.
Setting up the Confluence MCP server
Use Atlassian OAuth or an API token scoped to specific spaces. Confluence permissions are space-and-page-grained; grant the narrowest set, because internal wikis hold more sensitive detail than people remember.
Auth: Atlassian OAuth or API token, space-scoped. Credentials belong in the environment or a secret store — never in the repo or the agent's sandbox. Wiring a server into Claude Code or Cursor is the same for every tool; the step-by-step lives in the MCP connection guide.
The risk to watch with Confluence
Company wikis accumulate secrets, roadmaps, and personnel detail. Space-level scoping and audited access matter more here than the read-only/write distinction.
More broadly, a production-grade Confluence integration needs space-level scoping and keeping internal knowledge access audited. The MCP security checklist applies to every integration — we build servers with these controls baked in.
Troubleshooting
If the Confluence tools don't appear after setup, it's almost always auth or transport. See MCP server not connecting for the step-by-step fix — and note that hosted servers often need OAuth, not a plain API key. To understand how MCP relates to ordinary tool use, see MCP vs function calling.
Frequently asked questions
Is there an official MCP server for Confluence?
Community MCP servers exist. Whichever you use, a production setup needs space-level scoping and keeping internal knowledge access audited.
How does authentication work for Confluence over MCP?
Atlassian OAuth or API token, space-scoped. Credentials should never live in the sandbox or the repo; route them through your client's secret handling or a vaulted credential.
What can an agent actually do with Confluence?
Have an agent read design docs for context; Generate documentation pages from a change; Search internal knowledge while working. Start read-only and add write access deliberately, behind audit logging.
Is it safe to give agents access to Confluence?
Yes, when scoped correctly: least-privilege credentials, read-only by default, audit logs on every call, and human approval for any high-impact action. Space-level scoping and keeping internal knowledge access audited.