MCP.so
Sign In
Servers

MCP Atlassian Server

@petrsovadina

MCP server for integrating Atlassian products (Confluence, Jira) with Model Context Protocol

Overview

What is MCP Atlassian Server?

MCP Atlassian Server is a Model Context Protocol server that integrates with Atlassian products (Confluence and Jira). It enables direct access to Confluence content and Jira tickets through the MCP interface, allowing AI assistants to query and retrieve information from these Atlassian tools.

How to use MCP Atlassian Server?

Install via Smithery (npx -y @smithery/cli install @petrsovadina/mcp-atlassian --client claude) or manually (clone repo, npm install, npm run build). Configure environment variables for Confluence and Jira (URL, username, API token). Add the server configuration to your MCP settings file with the appropriate credentials.

Key features of MCP Atlassian Server

  • Search Confluence content using CQL
  • Access Confluence pages, attachments, and comments
  • Filter Confluence results by space
  • Search Jira issues using JQL
  • Retrieve Jira issue details including status, assignee, timestamps
  • Optional result limits (1‑50 per query)

Use cases of MCP Atlassian Server

  • Search for recent Confluence pages in a specific space
  • Find Jira issues assigned to a team under a particular project
  • Retrieve detailed information about a single Confluence page or Jira issue via resource templates
  • Filter and list open Jira issues by project
  • Query knowledge base documentation across Confluence spaces

FAQ from MCP Atlassian Server

What environment variables are required?

For Confluence: CONFLUENCE_URL, CONFLUENCE_USERNAME, CONFLUENCE_API_TOKEN. For Jira: JIRA_URL, JIRA_USERNAME, JIRA_API_TOKEN.

What tools does the server expose?

Two tools: confluence_search (accepts a CQL query, optional limit) and jira_search (accepts a JQL query, optional fields and limit).

How are resources identified?

Confluence pages use the template confluence://{space_key}/pages/{title}. Jira issues use jira://{project_key}/issues/{issue_key}.

What authentication is needed?

Both Confluence and Jira require an API token set via environment variables. There is no OAuth or user‑interactive flow.

Are there any query limits?

Yes, both search tools accept an optional limit parameter. The valid range is between 1 and 50 results.

More from Version Control