Atlassian Integration MCP Server

Developer ToolsPython

Connect to Atlassian tools like Confluence and Jira

Available Tools

jira_search_issues

Search for Jira issues using JQL (Jira Query Language)

jqlmax_resultsfields

jira_get_issue

Get detailed information about a specific Jira issue

issue_keyfields

confluence_search

Search for content in Confluence

queryspace_keylimit

confluence_get_page

Get content of a specific Confluence page

page_idexpand

The Atlassian Integration MCP provides seamless access to Atlassian's suite of collaboration tools, specifically Confluence and Jira. It enables AI assistants to retrieve information from your knowledge base, create and update issues, and interact with your team's workflow directly through the Atlassian API. With support for both OAuth 2.0 and Personal Access Token authentication, this MCP works with Atlassian Cloud, Server, and Data Center deployments. It's designed to enhance productivity by allowing AI assistants to access your team's documentation and project management tools in a secure and context-aware manner.

Overview

The Atlassian Integration MCP allows AI assistants to interact with Confluence and Jira, providing access to your team's knowledge base and project management tools. This integration enables AI to search, retrieve, and potentially update content in your Atlassian environment.

Installation

Prerequisites

Python 3.8 or higher
Access to Atlassian Confluence and/or Jira instances
Appropriate authentication credentials (OAuth 2.0 tokens for Cloud or Personal Access Tokens)

Installation Methods

Using pip

pip install mcp-atlassian

Using Docker

docker pull ghcr.io/sooperset/mcp-atlassian:latest

Configuration

Environment Variables

Configure the MCP using environment variables:

For Jira:

JIRA_URL: URL of your Jira instance
JIRA_USERNAME: Your Jira username (for PAT authentication)
JIRA_API_TOKEN: Your Jira API token (for PAT authentication)

For Confluence:

CONFLUENCE_URL: URL of your Confluence instance
CONFLUENCE_USERNAME: Your Confluence username (for PAT authentication)
CONFLUENCE_API_TOKEN: Your Confluence API token (for PAT authentication)

For OAuth 2.0 (Atlassian Cloud):

ATLASSIAN_OAUTH_CLIENT_ID: OAuth client ID
ATLASSIAN_OAUTH_CLIENT_SECRET: OAuth client secret
ATLASSIAN_OAUTH_REFRESH_TOKEN: OAuth refresh token

Authentication Setup

OAuth 2.0 (Recommended for Atlassian Cloud)

Create an OAuth 2.0 app in the Atlassian Developer Console
Configure the callback URL (typically http://localhost:8000/callback)
Run the OAuth authorization script:
```
python -m scripts.oauth_authorize
```
Follow the browser prompts to authorize the application
The script will output the necessary tokens to add to your environment

Personal Access Token (PAT)

Generate a PAT from your Atlassian account settings
Set the appropriate environment variables with your username and token

Running the MCP

Command Line

python -m mcp_atlassian

Docker

docker run -p 8000:8000 \
  -e JIRA_URL=https://your-instance.atlassian.net \
  -e CONFLUENCE_URL=https://your-instance.atlassian.net \
  -e ATLASSIAN_OAUTH_CLIENT_ID=your-client-id \
  -e ATLASSIAN_OAUTH_CLIENT_SECRET=your-client-secret \
  -e ATLASSIAN_OAUTH_REFRESH_TOKEN=your-refresh-token \
  ghcr.io/sooperset/mcp-atlassian:latest

Multi-User Authentication

The MCP supports multi-user authentication for HTTP-based transports:

For OAuth: Include Authorization: Bearer <oauth_token> in your requests
For PAT: Include Authorization: Token <personal_access_token> in your requests

This allows AI assistants to interact with Atlassian tools using individual user permissions.

Troubleshooting

Ensure your environment variables are correctly set
Check network connectivity to your Atlassian instance
Verify authentication credentials are valid and have appropriate permissions
For OAuth issues, try re-running the authorization script to refresh tokens