Log inSubmit server
Confluence logo

Confluence

Read, search, organize, and create Confluence Cloud pages through MCP tools backed by OAuth.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Confluence

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "confluence": {
      "command": "npx",
      "args": [
        "-y",
        "@pipeworx/mcp-confluence"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

The Confluence MCP server connects AI clients to the Confluence Cloud REST API v2 through OAuth. It can discover spaces, list and read pages, search by keyword or CQL, and create pages under a parent or space. Use the hosted Pipeworx endpoint for gateway access, or run the package locally over stdio without a Pipeworx account.

Use cases

  • Search Confluence documentation by keyword or CQL
  • Read a page's full content by ID
  • List pages within an engineering or sales space
  • Discover available Confluence spaces
  • Create a new page under a parent or space

Key features

  • Confluence Cloud REST API v2 access
  • OAuth-based account connection
  • Keyword and CQL page search
  • Page and space discovery
  • Page creation with parent or space targeting

Tools5

What Confluence lets your agent do. Expand a tool for its parameters.

Self-reported~176 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

7/7 checks healthy over the last 45d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 14d ago via OSV.dev · @pipeworx/mcp-confluence (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

npm downloads430Package downloads in the last 30 days.
Last commit1d agoMost recent push to the default branch.
Tools exposed5Callable tools this server registers over MCP.

How Confluence works

What the Confluence MCP server does

The Confluence MCP server exposes selected Confluence Cloud REST API v2 operations through the Model Context Protocol. It supports both reading existing documentation and creating new pages. Agents can discover available spaces, inspect page metadata, retrieve full page content, search across pages, and publish a new page.

This fits workflows where an AI client needs to answer questions from Confluence, locate documentation by space, summarize page content, or create a new knowledge-base entry. The integration is described as an independent, unofficial connection and is not affiliated with or endorsed by Confluence's upstream provider.

How it works

The hosted option uses the Pipeworx gateway at https://gateway.pipeworx.io/confluence/mcp. The gateway connects to a Confluence account and exposes the pack's tools through MCP. Its connection also lists shared Pipeworx meta-tools, so the tool list on the hosted endpoint is larger than the five Confluence-specific tools. The gateway's ask_pipeworx function can route natural-language questions across the broader Pipeworx catalog.

A local option runs the npm package with npx and communicates over standard input and output. In local mode, the server responds to MCP initialization, tool-listing, and tool-call requests for the Confluence tools only; it does not add the gateway's shared meta-tools.

Setup and configuration

For hosted use, sign in to Pipeworx, connect a Confluence account, and configure an MCP client with the Confluence gateway URL. Direct HTTP callers can authenticate with a Pipeworx key using a bearer token.

For local use, add an MCP server entry with the npx -y @pipeworx/mcp-confluence command and no gateway account. The Confluence MCP server can also be started directly with that command to verify that it launches. The README names Claude Desktop, Cursor, and Windsurf as example clients.

Tools and capabilities

  • confluence_list_spaces discovers spaces and returns each space's ID, key, name, type, and status.
  • confluence_list_pages lists pages in a specified space, including IDs, titles, status, and version information.
  • confluence_get_page retrieves a page by ID with its title, body, status, version, and space details.
  • confluence_search finds pages using a keyword or CQL query and returns matching metadata with content excerpts.
  • confluence_create_page creates a page with a title and content under a parent page or in a selected space, returning the new page ID and URL.

Limitations and notes

The Confluence MCP server is focused on the operations documented above. Local mode avoids the Pipeworx gateway but is limited to this pack's tools. Hosted mode requires a connected Confluence account and Pipeworx authentication. This integration belongs to the Workplace & Productivity category.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Confluence

Confluence is an MCP server that connects AI clients to the Confluence Cloud REST API v2 through OAuth. Its main tools list spaces and pages, retrieve page content, search with keywords or CQL, and create pages under a parent page or space. It can run through Pipeworx or locally over stdio.

Install it locally by configuring your MCP client to run `npx -y @pipeworx/mcp-confluence`. This uses stdio and does not require a Pipeworx account. Alternatively, configure the hosted MCP URL `https://gateway.pipeworx.io/confluence/mcp` after connecting Confluence in Pipeworx.

Add either the hosted Pipeworx MCP URL or the local `npx` command to Cursor's MCP configuration, depending on whether you want gateway access or a standalone stdio connection.

Yes, `confluence_create_page` creates a page with a title and content. You can place it under a parent page ID or specify a space key, and the tool returns the resulting page ID and URL.

Related MCP servers

All in Workplace & Productivity All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewConfluence AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/confluence?style=directory)](https://allmcps.com/mcp/confluence)
HTML Embed
<a href="https://allmcps.com/mcp/confluence"><img src="https://allmcps.com/api/badge/confluence?style=directory" alt="Confluence on AllMCPs" /></a>