home / mcp / mcp atlassian mcp server

MCP Atlassian MCP Server

MCP server for Atlassian (Confluence & Jira)

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "samwang0723-mcp-atlassian": {
      "url": "http://localhost:3007/mcp",
      "headers": {
        "ATLASSIAN_HOST": "https://your-domain.atlassian.net",
        "ATLASSIAN_EMAIL": "[email protected]",
        "ATLASSIAN_API_TOKEN": "your-api-token"
      }
    }
  }
}

You have an MCP (Model Context Protocol) server designed to give AI agents rich, structured access to Atlassian products such as Confluence and Jira. It exposes a comprehensive toolset that lets you search, manage content, spaces, pages, comments, and labels in Confluence, as well as create, update, and track Jira issues and projects. This makes it easy to integrate Atlassian workflows into your AI agents with robust safety, pagination, and error handling.

How to use

You connect an MCP client to the server URL exposed by the MCP host. The server provides a standard HTTP MCP endpoint that accepts requests for Confluence and Jira operations. You can start with a simple search or content retrieval to verify connectivity, then expand to create and update actions as your trust and needs grow. Use the predefined tools to perform indexing, querying, and CRUD operations across Confluence spaces, pages, comments, labels, and Jira issues, transitions, and comments. The server supports read-only mode and tool filtering to keep your exploration safe while you learn the capabilities.

Typical usage patterns include:

How to install

Prerequisites you need before installing this MCP server are straightforward. You should have Node.js version 16 or higher and either npm or yarn for package management. You also need an Atlassian account with an API token to access Confluence and Jira. Docker is optional if you prefer containerized deployment.

Install options are available depending on your preferred setup. Choose one of the following workflows and follow the exact steps.

Standard installation steps you can follow in sequence are:

1. Clone the project repository to your development machine.

2. Install dependencies using npm or yarn.

3. Create a .env file at the project root and provide your Atlassian credentials.

4. Start the MCP server using the recommended start command for your environment.

Docker-based installation steps you can follow in sequence are:

1. Clone the repository.

2. Create a .env file with your Atlassian credentials.

3. Build the Docker image and run the container, or use Docker Compose to orchestrate the service.

# Basic standard installation (live commands shown as examples)
# 1) Clone the repository
git clone https://github.com/yourusername/mcp-atlassian.git
cd mcp-atlassian

# 2) Install dependencies
npm install

# 3) Create environment file
# ATLASSIAN_HOST=https://your-domain.atlassian.net
# [email protected]
# ATLASSIAN_API_TOKEN=your-api-token

# 4) Start the server
npm start

Configuration example

{
  "mcp-searchapi": {
    "name": "mcp-searchapi",
    "type": "streamable-http",
    "streamable": true,
    "url": "http://localhost:3007/mcp"
  }
}

Available tools

search_confluence

Search Confluence content using v1 API with CQL for powerful content discovery.

search_confluence_pages_by_title

Search pages by title using the v2 API for precise title-based results.

get_confluence_space_by_id_or_key

Retrieve information about a specific Confluence space by its ID or key.

get_confluence_spaces

List all available Confluence spaces with optional pagination.

get_confluence_content

Fetch the content of a specific Confluence page by its ID with optional body format.

get_confluence_pages

List all pages within a given Confluence space.

get_confluence_child_pages

Retrieve child pages for a specified page.

confluence_create_page

Create a new Confluence page with title, content, and optional metadata.

confluence_update_page

Update an existing Confluence page including title, content, and versioning.

update_confluence_page_title

Update only the title of a Confluence page.

confluence_delete_page

Delete a Confluence page by its ID.

get_confluence_pages_by_label

Find pages that have a specific label (with optional space scoping and pagination).

get_confluence_page_labels

Retrieve labels for a specific Confluence page.

add_confluence_page_labels

Add one or more labels to a Confluence page.

get_confluence_page_comments

Get regular comments attached to a Confluence page.

add_confluence_page_comment

Add a new comment to a Confluence page.

get_confluence_page_inline_comments

Fetch inline comments on a Confluence page.

create_confluence_footer_comment

Create a footer comment associated with a Confluence page or related blog post.

search_jira_issues

Search Jira issues using JQL with optional max results.

get_jira_issue

Get detailed information about a Jira issue by its key.

jira_get_all_projects

List all accessible Jira projects.

jira_create_issue

Create a new Jira issue with required fields and optional additional fields.

jira_update_issue

Update fields of an existing Jira issue.

jira_add_comment

Add a comment to a Jira issue.

jira_get_transitions

Get available transitions for a Jira issue.

jira_transition_issue

Transition a Jira issue to a new status with an optional comment.