Atlassian MCP Server

20 MCP tools for Jira issues, Confluence pages, and cross-product search — pure Python, no Node.js dependency.

Overview

The Atlassian MCP Server provides 20 tools for interacting with Jira and Confluence. It was rewritten as a direct Python REST API client, eliminating the earlier Node.js dependency. The server handles issue tracking, sprint management, Confluence page retrieval, and cross-product search through a single unified interface.

Property Value
Port 50052
Transport JSON-RPC at /jsonrpc
Namespace f3iai
Backend Atlassian Cloud REST API

Tools

Jira — Issues

Tool Description
jira_search_issues Search issues using JQL (Jira Query Language)
jira_get_issue Get detailed issue information including fields, comments, and transitions
jira_create_issue Create a new issue in a project
jira_update_issue Update fields on an existing issue
jira_transition_issue Move an issue to a new status (e.g., In Progress, Done)
jira_add_comment Add a comment to an issue
jira_list_issue_comments List all comments on an issue
jira_get_issue_transitions Get available status transitions for an issue

Jira — Projects and Boards

Tool Description
jira_list_projects List all accessible Jira projects
jira_get_project Get project details including issue types and workflows
jira_list_sprints List sprints in a board, filtered by state (active, future, closed)
jira_get_sprint_issues Get issues in a specific sprint

Confluence — Pages

Tool Description
confluence_get_page Get a Confluence page by ID, including body content
confluence_get_page_children List child pages of a given page
confluence_search Search Confluence using CQL (Confluence Query Language)
confluence_list_spaces List all Confluence spaces
confluence_get_space Get details of a specific space
confluence_create_page Create a new Confluence page
confluence_update_page Update an existing Confluence page

Cross-Product

Tool Description
atlassian_search Unified search across both Jira and Confluence

Common Use Cases

  • “Show open issues in the FF project” — calls jira_search_issues with JQL project = FF AND status != Done
  • “What’s in sprint 42?” — calls jira_get_sprint_issues
  • “Find the FFO MCP Server spec in Confluence” — calls confluence_search
  • “Get the TypeDB deployment page” — calls confluence_get_page with page ID
  • “Create a bug for the Compass API” — calls jira_create_issue
  • “Move FF-279 to Done” — calls jira_transition_issue

Authentication

The server authenticates to Atlassian Cloud using an API token:

Environment Variable Description
ATLASSIAN_URL Atlassian instance URL (e.g., https://eupraxialabs.atlassian.net)
ATLASSIAN_EMAIL Account email for authentication
ATLASSIAN_API_TOKEN API token generated from Atlassian account settings
ATLASSIAN_CLOUD_ID Cloud instance ID (used for some API endpoints)

Health Endpoints

Endpoint Purpose
/health Liveness — process is running
/ready Readiness — can authenticate to Atlassian API