Atlassian

Provides MCP access to Jira and Confluence data and actions via tokens and environment-based configuration.

other

3.7k

GitHub Stars

other

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "sooperset-mcp-atlassian": {
      "command": "uvx",
      "args": [
        "mcp-atlassian"
      ],
      "env": {
        "JIRA_URL": "https://your-company.atlassian.net",
        "JIRA_USERNAME": "user@example.com",
        "CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
        "JIRA_API_TOKEN": "YOUR_API_TOKEN",
        "CONFLUENCE_USERNAME": "user@example.com",
        "CONFLUENCE_API_TOKEN": "YOUR_API_TOKEN"
      }
    }
  }
}

Model Context Protocol (MCP) Atlassian enables AI assistants to interact with Jira and Confluence across Cloud and Server deployments. You can search content, create and update pages, manage issues, and automate workflows from your IDE or CI, with flexible authentication, proxying, and multi-user support.

How to use

You connect your MCP Atlassian server to your AI assistant and start issuing high-level tasks to automate Jira and Confluence work. Typical workflows include updating Jira from notes, searching Confluence content and summarizing it, filtering Jira issues by project and urgency, and creating or updating Confluence pages or Jira issues from your AI prompts. Start by configuring authentication, then run the MCP server in your preferred transport (stdio for local execution or HTTP for a persistent service). After that, point your MCP client in the IDE to the server URL and begin issuing actions such as searching projects, creating pages, or updating issues.

How to install

Prerequisites: you need Docker installed on your machine.

Pull the pre-built Docker image.

# Pull Pre-built Image
docker pull ghcr.io/sooperset/mcp-atlassian:latest

Additional sections

Configuration basics for MCP Atlassian include how you authenticate, how you run the server, and how you enable or restrict tools. The server supports multiple authentication methods, proxy configuration, custom HTTP headers, and multi-cloud OAuth scenarios. You can choose a simple API token flow for Cloud, a personal access token for Server/Data Center, or an OAuth 2.0 flow for Cloud. For advanced deployments, BYOT lets you bring your own OAuth access token and manage token refresh externally.

When you run the server, you choose the transport that best fits your workflow. Use SSE or streamable-http for HTTP-based transport, or run in stdio mode for a local, in-process instance. The server can operate in single-user mode with a fixed set of tokens, or in multi-user mode where each user provides their own OAuth or PAT tokens.

Security and troubleshooting tips are included to help you diagnose authentication issues, SSL verification, and permission problems. You can enable verbose logging to inspect request headers and verify that custom headers are applied correctly. If you need multi-cloud support, you can route requests by cloud instance using headers like X-Atlassian-Cloud-Id.

Available tools

jira_search

Search Jira issues using JQL and return matching results.

jira_get_issue

Retrieve details for a specific Jira issue.

jira_create_issue

Create a new Jira issue with given fields.

jira_update_issue

Update fields on an existing Jira issue.

jira_transition_issue

Change the status of a Jira issue.

jira_get_issue_sla

Calculate SLA metrics for Jira issues.

confluence_search

Search Confluence using CQL to find pages.

confluence_get_page

Retrieve the content of a Confluence page.

confluence_create_page

Create a new Confluence page.

confluence_update_page

Update content of an existing Confluence page.

confluence_add_comment

Add comments to a Confluence page.

confluence_get_page_views

Get page view statistics (Cloud only).

Built by

VeilStrat

AI signals for GTM teams