home / mcp / confluence cloud mcp server

Confluence Cloud MCP Server

A Model Context Protocol (MCP) server that enables AI assistants to manage Confluence Cloud spaces, pages, and content through a standardized interface.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "aaronsb-confluence-cloud-mcp": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/aaronsb/confluence-cloud-mcp:latest"
      ],
      "env": {
        "CONFLUENCE_EMAIL": "[email protected]",
        "CONFLUENCE_DOMAIN": "your-domain.atlassian.net",
        "CONFLUENCE_API_TOKEN": "your-api-token"
      }
    }
  }
}

You deploy a MCP server that lets AI assistants interact with Confluence Cloud to manage spaces, pages, and content using a standardized, programmable interface. It exposes tools for listing spaces, reading and updating pages, converting content to Markdown, and performing searches and label management, enabling seamless automation and integration with MCP-compatible assistants.

How to use

Connect to the MCP server from your MCP client or orchestration layer. You can enable several operations: list spaces, get space details, list pages in a space, read a page with its content converted to Markdown, create or update pages, search content with CQL, and manage page labels. The server exposes these tools under clear names, and the page content is automatically converted to Markdown for easier downstream processing.

How to install

Prerequisites you need before starting are: Node.js (for local builds and scripts) and Docker if you plan to run the server as a container. You will also need a Confluence Cloud account with an API token, your email, and your Confluence domain.

# Option 1. Run with Docker (recommended)
docker run --rm -i \
  -e CONFLUENCE_API_TOKEN=your-api-token \
  -e [email protected] \
  -e CONFLUENCE_DOMAIN=your-domain.atlassian.net \
  ghcr.io/aaronsb/confluence-cloud-mcp:latest

If you prefer building locally from source, follow these steps to install dependencies, build, and run the server.

# Option 2. Building Locally
# 1) Clone the repository
# git clone https://github.com/aaronsb/confluence-cloud-mcp.git
# cd confluence-cloud-mcp

# 2) Build the Docker image locally (uses the project's build scripts)
./scripts/build-local.sh

# 3) Run the server with credentials
CONFLUENCE_API_TOKEN=your-token CONFLUENCE_EMAIL=your-email ./scripts/run-local.sh

If you want to run directly with npx from GitHub, you can pass your credentials via an environment file.

# Option 3. Using npx (directly from GitHub)

# With inline environment variables
CONFLUENCE_DOMAIN=your-domain.atlassian.net \
[email protected] \
CONFLUENCE_API_TOKEN=your-api-token \
npx github:aaronsb/confluence-cloud-mcp

# Or with an environment file
npx github:aaronsb/confluence-cloud-mcp --env /path/to/.env

# .env should contain
CONFLUENCE_DOMAIN=your-domain.atlassian.net
[email protected]
CONFLUENCE_API_TOKEN=your-api-token

If you prefer installing from source, follow these steps to install dependencies, configure credentials, build, and run.

# Option 4. From Source

# 1) Install dependencies
npm install

# 2) Configure environment variables by creating a .env file
# CONFLUENCE_DOMAIN=your-domain.atlassian.net
# [email protected]
# CONFLUENCE_API_TOKEN=your-api-token

# 3) Build the project
npm run build

# 4) Run the server
node build/index.js

Additional notes

The server is designed to work with MCP-compatible clients. You can integrate it into your automation workflows by adding it to your MCP configuration using either a local npx-based command or a Docker-based runtime. Each approach requires your Confluence credentials and the domain you use to access Confluence Cloud.

Security reminder: treat your API token as a secret. Do not share your credentials in public repositories or logs. Use environment files or secret management where possible.

Tools and usage concepts

The server provides a set of focused tools to interact with Confluence Cloud programmatically. These are designed to be invoked by MCP-compatible clients to automate common tasks.

Available tools

list_confluence_spaces

List all spaces in Confluence Cloud.

get_confluence_space

Get details about a specific Confluence space.

list_confluence_pages

List pages within a Confluence space.

get_confluence_page

Get a specific page and its content, with content converted to Markdown.

create_confluence_page

Create a new page within a Confluence space.

update_confluence_page

Update an existing Confluence page.

search_confluence_pages

Search Confluence content using CQL.

get_confluence_labels

Retrieve labels for a page.

add_confluence_label

Add a label to a page.

remove_confluence_label

Remove a label from a page.