# Atlassian MCP for AI Agents MCP > The Atlassian Jira & Confluence MCP lets your AI agent operate across your entire project ecosystem. Instead of jumping between dashboards, you can ask natural language questions to audit complex Jira issues, review active agile sprints, and retrieve deep knowledge from enterprise Confluence wikis—all in one conversation. ## Overview - **Category:** industry-titans - **Price:** Free - **Tags:** agile-boards, issue-tracking, wiki, sprint-management, bug-tracking, team-collaboration ## Description Managing projects means juggling tasks, bugs, requirements, and documentation. These pieces live across different systems: your development board in Jira, and your spec sheets in Confluence. Before this MCP, you'd spend time copying ticket numbers into search queries or manually cross-referencing dates between two separate dashboards. With this integration, your AI agent acts like a direct navigator across both platforms. You can ask it to find all high-priority tickets related to a specific feature and then immediately pull the associated architectural decision document from Confluence—all without leaving your chat window. It turns complex, siloed workflows into simple, conversational commands. This entire catalog of tools is hosted on Vinkius, meaning you connect once and get access to this deep project intelligence alongside thousands of other services. ## Tools ### list_projects Lists every project key available in your Jira instance so you know which domain to query. ### search_issues Searches all Jira issues using complex JQL queries based on criteria like assignee or status. ### get_issue Fetches the complete details for a single Jira issue when you know its exact key. ### list_boards Shows all active agile boards, helping you decide where to look for backlogs or running sprints. ### list_sprints Retrieves a list of historical and current sprints tied to a specific Jira board. ### get_myself Pulls information about the user who is currently authenticated to Atlassian, confirming permissions and identity. ### list_spaces Lists all available spaces in Confluence so you know which knowledge area to search. ### search_content Searches across your entire Confluence wiki using CQL, finding pages related to a topic or keyword. ### get_page Extracts the full rich text content from a specific Confluence page for detailed reading. ## Prompt Examples **Prompt:** ``` What are all high-priority tickets assigned to me related to login errors? ``` **Response:** ``` **🔎 Sprint Audit: Login Errors** I found 4 critical issues currently in progress: * `ENG-501`: Password Reset flow failure (Priority: Blocker) * `ENG-489`: MFA enrollment bug (Priority: High) * `ENG-320`: Session timeout error (Priority: Medium) These tickets are linked to the 'Authentication Flow' Confluence page. Should I pull the full text of that spec for context? ``` **Prompt:** ``` Find documentation on our new billing API structure. ``` **Response:** ``` I searched across all available spaces and found two pages: **Payment Gateway v2** (most recent) and *Legacy Stripe Migration Docs*. The first one seems most relevant. It details rate limits, required fields, and the full JSON schema for endpoints like `/v2/charge`. Would you like me to read out the text from this page now? ``` **Prompt:** ``` List all active boards and see what's due next week. ``` **Response:** ``` **📊 Board Overview** You have 5 active Jira Boards. Here’s a quick look at immediate deadlines: * `CORE-BACKEND`: Sprint 44 has 12 tickets due next Tuesday. * `MOBILE-REVAMP`: Sprint 12 is clear, but there are 8 stories flagged for review by end of week. The `CORE-BACKEND` board looks like the priority right now. Want me to pull a list of those specific tickets? ``` ## Capabilities ### Search Jira issues using advanced criteria Find specific tickets or groups of issues across your organization using complex JQL queries. ### Review project status and active sprints Get an overview of all available agile boards, list current sprints, and track which tasks are due in the near future. ### Access deep documentation from Confluence Search across your entire wiki knowledge base using CQL and extract the full text content of specific pages or spaces. ### Understand user identities within Atlassian Automatically pull current authenticated user information to ensure actions are assigned correctly. ## Use Cases ### The requirement changes mid-sprint A developer realizes a bug ticket (ENG-402) needs more info. They ask their agent, which uses get_page to pull the detailed architectural design from Confluence and feeds it back into the conversation alongside the issue details. ### Auditing project scope before a meeting A Product Manager asks the agent for all active boards (list_boards) and then uses search_issues to aggregate every open story related to 'checkout flow' across those boards, giving them one cohesive report. ### Finding outdated technical guides A support specialist needs a guide for an old API. They prompt the agent, which uses search_content on Confluence and then pulls the full text of the most relevant page (get_page), allowing them to answer the customer instantly. ### Getting status across multiple teams A Scrum Master asks for all active sprints in a department. The agent uses list_sprints and reports back which boards are running, allowing the master to see overall project health at a glance. ## Benefits - Get instant visibility into project status. Use list_boards and list_sprints to quickly see which agile boards have active sprints without logging into the dashboard. - Tackle complex debugging cycles. Combine search_issues with get_page to find a bug ticket, then immediately pull the original specification document that describes the required behavior. - Stop manual data gathering. Instead of copying criteria, your agent can summarize complete epics or audit sprints across multiple boards using list_sprints and search_issues. - Know exactly who's working on what. The get_myself tool confirms current user identity, ensuring that all ticket actions are attributed correctly from the start. - Rapid knowledge retrieval. Use search_content to find technical guides or architectural patterns buried deep in Confluence wikis, then use get_page to stream the full text for context. ## How It Works The bottom line is you use your agent to talk to both Atlassian services simultaneously from one chat interface. 1. Subscribe to this secure MCP and provide your Atlassian Domain, Email, and API Token. 2. Connect the credentials to your preferred AI client (Claude, Cursor, Windsurf, etc.). 3. Use natural language prompts to ask questions that cross-reference tasks in Jira with documentation in Confluence. ## Frequently Asked Questions **How does the Atlassian Jira & Confluence MCP help with project planning?** This MCP gives your agent real-time access to both task boards and documentation. You can ask it to audit a sprint's scope by checking related tickets in Jira against the official requirements documented in Confluence, giving you full context before starting. **Can I use this MCP to find API details for old features?** Yes. If the documentation exists in Confluence, your agent can search across all spaces and pull up the specific page content using tools like get_page. It’s perfect for finding archived or legacy technical specs. **Does the Atlassian MCP support different types of Jira boards?** It supports multiple board types, allowing your agent to list and view active sprints across various project domains. This means you get a comprehensive overview of all work happening in one place. **What if I need to cross-reference an issue key with a document?** You just ask the agent directly. It can use your issue key to find the ticket details and then automatically search Confluence for related documentation or specifications, linking all the information together. **Is this Atlassian MCP better than using a separate Jira integration?** Because it handles both Jira and Confluence in one flow, you avoid having to use two different tools. You keep your conversation focused on solving problems rather than managing multiple connections.