Skip to main content
The Atlassian MCP server provides comprehensive access to Jira and Confluence, enabling AI agents to manage issues, sprints, boards, epics, and documentation pages.

Overview

This connector provides 51 tools covering:
  • Jira Issues - Create, update, search, and transition issues
  • Sprints & Boards - Manage agile workflows
  • Epics & Backlogs - Organize work hierarchically
  • Time Tracking - Log and manage work time
  • Confluence - Create and edit documentation pages

Integration Options

Option A: Connect to Atlassian’s Remote MCP Server

Atlassian provides an official hosted MCP server. Connect using the Vercel AI SDK or Pydantic AI.
import { experimental_createMCPClient } from "ai/mcp";
import { createHeroku } from "@heroku/ai-provider";

const mcpClient = await experimental_createMCPClient({
  transport: {
    type: "sse",
    url: "https://mcp.atlassian.com/v1/sse",
  },
});

const tools = await mcpClient.tools();
const heroku = createHeroku();

const response = await generateText({
  model: heroku("claude-4-sonnet"),
  tools,
  prompt: "List my open Jira issues assigned to me",
});
Atlassian’s hosted MCP server handles OAuth authentication automatically when you connect.

Option B: Deploy to Heroku

Deploy a community MCP server to Heroku for full control over configuration: Deploy to Heroku After deployment, configure your Atlassian credentials:
heroku config:set JIRA_HOST=https://your-domain.atlassian.net -a your-app-name
heroku config:set JIRA_EMAIL=your-email@example.com -a your-app-name
heroku config:set JIRA_API_TOKEN=your-api-token -a your-app-name
Get your API token from Atlassian Account Settings.

Register with Heroku AI

heroku ai:mcp:servers:add jira-mcp \
  --app your-inference-app \
  --server-app your-jira-mcp-app

Available Tools

Jira Issue Management

ToolDescription
jira_search_issuesSearch issues using JQL (Jira Query Language)
jira_get_issueGet detailed information about a specific issue
jira_create_issueCreate a new issue with type, summary, description
jira_update_issueUpdate issue fields (summary, description, assignee, etc.)
jira_delete_issueDelete an issue
jira_transition_issueMove issue through workflow (To Do → In Progress → Done)
jira_link_issuesCreate links between related issues
jira_add_commentAdd a comment to an issue
jira_get_commentsRetrieve all comments on an issue

Sprint Management

ToolDescription
jira_list_sprintsList all sprints for a board
jira_create_sprintCreate a new sprint
jira_start_sprintStart a sprint with goal and dates
jira_close_sprintComplete a sprint
jira_move_issue_to_sprintAdd issues to a sprint

Board & Backlog

ToolDescription
jira_list_boardsList all boards in the project
jira_create_boardCreate a new Scrum or Kanban board
jira_get_board_issuesGet all issues on a board
jira_get_backlogGet backlog items
jira_move_issue_to_backlogMove issue to backlog
jira_rank_issuesReorder issues in backlog

Epics & Versions

ToolDescription
jira_create_epicCreate a new epic
jira_assign_issue_to_epicAdd issues to an epic
jira_get_epic_issuesList all issues in an epic
jira_list_versionsList project versions/releases
jira_create_versionCreate a new version
jira_update_versionUpdate version details
jira_delete_versionDelete a version

Time Tracking

ToolDescription
jira_add_worklogLog time spent on an issue
jira_update_worklogUpdate existing work log
jira_delete_worklogRemove a work log entry
jira_get_worklogsGet all work logs for an issue

Confluence Pages

ToolDescription
confluence_list_pagesList pages in a space
confluence_get_pageGet page content
confluence_get_page_childrenGet child pages
confluence_create_pageCreate a new page
confluence_update_pageUpdate page content
confluence_delete_pageDelete a page
confluence_searchSearch using CQL (Confluence Query Language)

Confluence Collaboration

ToolDescription
confluence_list_spacesList all spaces
confluence_get_commentsGet page comments
confluence_add_commentAdd a comment to a page
confluence_get_labelsGet page labels
confluence_add_labelAdd a label to a page
confluence_search_userSearch for users

Using with Heroku AI Agents

Once registered, use the connector through the Heroku Agents API:
import os
from openai import OpenAI

client = OpenAI(
    base_url=os.getenv("INFERENCE_URL") + "/v1",
    api_key=os.getenv("INFERENCE_KEY")
)

response = client.chat.completions.create(
    model=os.getenv("INFERENCE_MODEL_ID"),
    messages=[
        {"role": "user", "content": "Create a bug ticket for the login page crash on mobile devices"}
    ],
    extra_body={
        "heroku": {
            "mcp_servers": ["jira-mcp"]
        }
    }
)

print(response.choices[0].message.content)

Configuration

VariableDescriptionRequired
JIRA_HOSTYour Atlassian instance URL (e.g., https://company.atlassian.net)Yes
JIRA_EMAILEmail associated with your Atlassian accountYes
JIRA_API_TOKENAPI token from Atlassian account settingsYes
The same credentials work for both Jira and Confluence if they’re on the same Atlassian instance.

Example Use Cases

Sprint Planning

"Move all unfinished issues from Sprint 5 to Sprint 6"

Bug Triage

"Find all critical bugs created this week and assign them to the platform team"

Documentation

"Create a Confluence page documenting the new authentication flow based on the Jira epic PROJ-123"

Status Reports

"Summarize the progress of Sprint 7 including completed issues and remaining work"

Issue Creation

"Create a story for implementing dark mode with acceptance criteria and link it to the UI Refresh epic"

Jira Editions

Sprint and board management tools require Jira Software. If you’re using Jira Work Management or Jira Service Management, some agile-specific tools may not be available.

Additional Resources