Provides comprehensive integration with Atlassian Cloud products, enabling document management, search, content export, user activity tracking, and project management across both Confluence and Jira platforms.
Allows reading, creating, and updating pages; managing attachments and labels; searching content with CQL; navigating page hierarchies; exporting to HTML/Markdown; and tracking user activity and mentions across Confluence spaces.
Enables reading and searching issues with JQL; creating issues with custom fields; managing comments; listing projects, boards, and sprints; tracking time and user activity; and viewing personal dashboards with assigned and sprint tasks.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Atlassian Server search for open issues assigned to me in the current sprint"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Atlassian 제품(Confluence, Jira)과 통합하기 위한 Model Context Protocol (MCP) 서버입니다. AI 어시스턴트가 Atlassian Cloud API와 상호작용하여 문서 관리, 검색, 내보내기 기능을 사용할 수 있게 합니다.
Related MCP server: MCP Atlassian Server
읽기 & 검색: 페이지, 스페이스, 콘텐츠 접근
콘텐츠 관리: 페이지 생성, 수정, 댓글 작성
페이지 계층: 부모/자식 페이지 관계 탐색
내보내기: 이미지가 포함된 HTML 또는 Markdown으로 내보내기
첨부파일: 첨부파일 목록, 다운로드, 업로드
레이블: 페이지 레이블 관리
사용자: 사용자 검색 및 개인 활동 추적
개인 대시보드: 최근 페이지 및 멘션 확인
이슈: 이슈 읽기, 검색, 개인 작업 조회
프로젝트: 프로젝트 목록 및 탐색
보드 & 스프린트: 보드 목록, 스프린트 보기, 활성 작업 추적
댓글: 이슈에 댓글 추가
이슈 생성: 커스텀 필드를 포함한 새 이슈 생성
사용자 관리: 현재 사용자 정보 조회
개인 대시보드: 열린 이슈 및 스프린트 작업 확인
Atlassian 보안 설정으로 이동
"API 토큰 만들기" 클릭
토큰 라벨 입력 (예: "MCP Integration")
생성된 토큰을 복사 (한 번만 표시됩니다!)
⚠️ 중요: API 토큰은 비밀번호처럼 안전하게 보관하세요.
Claude Desktop
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json
Cursor
모든 OS: ~/.cursor/mcp.json
설정 파일을 열고 다음 내용을 추가하세요:
{
"mcpServers": {
"atlassian": {
"command": "npx",
"args": ["-y", "mcp-atlassian"],
"env": {
"ATLASSIAN_BASE_URL": "https://your-company.atlassian.net",
"ATLASSIAN_EMAIL": "your-email@company.com",
"ATLASSIAN_API_TOKEN": "YOUR_API_TOKEN_HERE"
}
}
}
}설정값 설명:
ATLASSIAN_BASE_URL: 회사의 Atlassian URL (끝에 / 제거)
ATLASSIAN_EMAIL: Atlassian 계정 이메일
ATLASSIAN_API_TOKEN: 1단계에서 생성한 API 토큰
Claude Desktop: 앱을 완전히 종료하고 다시 시작
Cursor: ⌘+Q (또는 Ctrl+Q) 후 다시 실행
AI 어시스턴트에게 다음과 같이 물어보세요:
"Atlassian MCP 서버가 연결되었는지 확인해줘"도구 | 설명 |
| 인증된 사용자 정보 조회 |
| 특정 사용자 정보 조회 |
| 사용자 활동으로 페이지 검색 |
| 사용자가 작성한 페이지 목록 |
| 사용자가 업로드한 첨부파일 목록 |
| ID 또는 제목으로 페이지 읽기 |
| CQL로 페이지 검색 |
| 접근 가능한 스페이스 목록 |
| 특정 스페이스 정보 조회 |
| 새 페이지 생성 |
| 기존 페이지 수정 |
| 하위 페이지 목록 |
| 상위 페이지 계층 조회 |
| 이미지 포함 HTML/Markdown 내보내기 |
| 페이지 첨부파일 목록 |
| 특정 첨부파일 다운로드 |
| 페이지에 파일 업로드 |
| 모든 콘텐츠와 함께 페이지 다운로드 |
| 페이지에 댓글 추가 |
| 페이지 레이블 조회 |
| 페이지에 레이블 추가 |
| 사용자 검색 |
| 내 최근 페이지 목록 |
| 나를 멘션한 페이지 찾기 |
도구 | 설명 |
| 인증된 사용자 정보 조회 |
| 특정 사용자 정보 조회 |
| 사용자 관련 이슈 검색 |
| 역할별 사용자 이슈 목록 (날짜 필터링) |
| 댓글 및 상태 변경 포함 사용자 활동 추적 |
| 시간 추적 항목 및 합계 조회 |
| 키로 이슈 상세 정보 읽기 |
| JQL로 이슈 검색 |
| 접근 가능한 프로젝트 목록 |
| 새 이슈 생성 |
| 이슈에 댓글 추가 |
| 스크럼/칸반 보드 목록 |
| 보드의 스프린트 목록 |
| 스프린트 상세 정보 조회 |
| 활성 스프린트의 내 작업 조회 |
| 모든 미해결 이슈 조회 |
"내가 작성한 Confluence 페이지 중 최근 5개를 보여줘""'API 문서'라는 제목이 포함된 페이지를 검색해줘""페이지 ID 882573681의 내용을 마크다운으로 보여줘""나한테 할당된 미해결 이슈를 모두 보여줘""현재 스프린트에서 내 작업 목록을 보여줘""접근 가능한 Jira 프로젝트 목록을 보여줘""스크럼 보드 목록을 보여줘""type = page AND creator = currentUser() 조건으로 Confluence 페이지를 검색해줘"API 토큰 재확인
Base URL 형식: https://your-company.atlassian.net (끝에 / 제거)
이메일과 토큰이 정확한지 확인
토큰에 필요한 권한이 있는지 확인
앱 완전 재시작
Claude Desktop: 앱을 완전히 종료하고 다시 시작
Cursor: ⌘+Q 후 재실행
설정 파일 확인
JSON 형식이 올바른지 확인 (쉼표, 중괄호 등)
환경 변수 이름이 정확한지 확인
검색 범위를 좁히면 속도가 향상됩니다:
"최대 10개의 결과만 보여줘"
"DEV 스페이스에서만 검색해줘"
"최근 1주일 이내의 이슈만 보여줘"GitHub 저장소에 이슈를 생성해주세요: Issues
Atlassian API 문서: Developer Docs
Atlassian Cloud만 지원 (Server/Data Center 미지원)
안전을 위해 삭제 작업은 지원하지 않음
일부 Confluence 매크로는 Markdown 변환 시 제한적
Atlassian Cloud API 속도 제한이 적용됨
MIT License
Made with ❤️ for better Atlassian integration
Adds a comment to a Confluence page. Can also be used to reply to an existing comment.
Inputs: pageId (required), content (required), parentCommentId.
Adds one or more labels to a Confluence page. Labels are useful for organizing and categorizing content.
Inputs: labels (required), pageId (required).
Adds a comment to an existing Jira issue. You can also control the visibility of the comment.
Inputs: body (required), issueKey (required), visibility.
Creates a new page or blog post in a Confluence space. Content can be provided in Markdown or Confluence storage format (XHTML).
Inputs: title (required), content (required), spaceKey (required), type, +1 more.
Creates a new issue in a Jira project. You must specify the project, issue type, and a summary. Other fields like description, priority, and assignee are optional.
Inputs: summary (required), issueType (required), projectKey (required), labels, +5 more.
Downloads a specific Confluence attachment and returns its content as a base64-encoded string. This is useful for reading the content of files attached to a page.
Inputs: attachmentId (required), version.
Exports a Confluence page to either HTML or Markdown format. All images in the page content are embedded directly into the exported file as base64 data, making it self-contained. Use this when you want to create a portable, self-contained file of a page.
Inputs: format (required), pageId (required).
Searches for Confluence users based on various criteria. This is useful for finding user details like account IDs when you have partial information or need to perform a broader search. Use this tool when you need to find users based on a query.
Inputs: cql, limit, start, expand, +3 more.
Get details of the authenticated Confluence user. Returns information about the current user including account ID, display name, email, and profile picture.
Inputs: no named parameters.
Search for pages that mention the current user. Returns pages where you have been @mentioned.
Inputs: limit, start, spaceKey.
Get details of a specific Confluence space by its key. Returns space information including name, type, status, and description.
Inputs: spaceKey (required), expand.
Get details for a specific Confluence user by a unique identifier like username, account ID, or email. Returns user profile information. Use this tool when you are confident you can uniquely identify a user.
Inputs: email, username, accountId.
Get details of the authenticated Jira user. Returns information about the current user including account ID, display name, email, and avatar URLs.
Inputs: no named parameters.
Get details for a specific Jira user by username, account ID, or email. Returns user profile information.
Inputs: email, username, accountId.
Get all issues assigned to you in the currently active sprint(s). Useful for daily standups or checking your current sprint workload. Can filter by specific board or project.
Inputs: boardId, projectKey.
List pages recently created or updated by the current user. Returns pages where you are the creator or last modifier.
Inputs: limit, start, spaceKey.
Get all unresolved issues currently assigned to you, organized by status and priority. Perfect for checking your backlog or identifying what needs attention.
Inputs: maxResults, projectKeys.
Performs a comprehensive download of a Confluence page, including its full content, metadata, and optionally, all of its attachments. Attachments are base64-encoded. Use this when you need the page content, metadata, and all associated attachments.
Inputs: pageId (required), attachmentTypes, maxAttachmentSize, includeAttachments.
Get comprehensive details about a specific sprint including its name, start/end dates, goals, and all issues included in the sprint.
Inputs: sprintId (required).
Get a stream of recent user activity, such as issue comments, status changes, and field updates. Can filter by activity type and project.
Inputs: days, startAt, username, accountId, +3 more.
Retrieve time tracking work logs (worklogs) for a specific user. Shows all time entries logged by the user within a date range, useful for timesheet and worklog reporting.
Inputs: endDate, startAt, username, accountId, +3 more.
List all accessible Agile boards (both Scrum and Kanban types). Can be filtered by project or board type to find specific boards.
Inputs: type, startAt, maxResults, projectKeyOrId.
Lists all attachments for a specific Confluence page. Can be filtered by filename or media type.
Inputs: pageId (required), limit, start, filename, +1 more.
List all attachments uploaded by a specific user, with optional filtering by space.
Inputs: limit, start, spaceKey, username, +1 more.
Retrieve the parent hierarchy of a Confluence page. Returns the full ancestry path from root to the immediate parent.
Inputs: pageId (required).
List child pages under a given Confluence page. Returns a hierarchical list of direct child pages.
Inputs: pageId (required), limit, start, expand.
Retrieves all labels for a specific Confluence page.
Inputs: pageId (required), limit, start, prefix.
Lists all Confluence spaces that the user has permission to view. Can be filtered by type and status.
Inputs: type, limit, status.
List issues where a user has a specific role (assignee, reporter, or creator). Includes date filtering for finding issues within specific time periods.
Inputs: role (required), endDate, startAt, username, +4 more.
Lists all Jira projects that the user has permission to view. This is useful for discovering available projects to work with.
Inputs: expand.
List all pages created by a specific user, with optional filtering by space or time range.
Inputs: limit, start, endDate, spaceKey, +3 more.
List all sprints for a specific Agile board. Can filter by sprint state (active, closed, or future) to find current or historical sprints.
Inputs: boardId (required), state, startAt, maxResults.
Retrieves the content of a Confluence page. You can specify the page by its ID or by its title and space key. The content can be returned in raw storage format (XHTML) or converted to Markdown. Use this for quickly reading the text content of a page.
Inputs: title, expand, format, pageId, +1 more.
Retrieves detailed information about a specific Jira issue, including its fields, status, and transitions. Use this to get the full picture of a single issue.
Inputs: issueKey (required), expand.
Performs a search for Confluence pages using Confluence Query Language (CQL). This is useful for finding pages that match specific criteria.
Inputs: cql (required), limit, start, expand.
Search for issues based on how a specific user is involved (assignee, reporter, creator, watcher, or any). Provides comprehensive filtering to find all issues a user is connected to.
Inputs: searchType (required), status, startAt, username, +4 more.
Searches for Jira issues using Jira Query Language (JQL). This is the primary way to find issues that match specific criteria.
Inputs: jql (required), fields, startAt, maxResults.
Search pages based on a user's involvement, such as being the creator or last modifier. Can filter by creator, last modifier, or both.
Inputs: searchType (required), limit, start, spaceKey, +2 more.
Updates an existing Confluence page. You must provide the page ID and its current version number to prevent conflicts. You can update the title, content, or both.
Inputs: pageId (required), version (required), title, content, +2 more.
Upload a file attachment to a Confluence page. The file should be provided as a base64-encoded string.
Inputs: file (required), pageId (required), filename (required), comment, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 40 tools
Most tools have clearly distinct purposes (Confluence vs Jira, different actions). However, a few pairs like 'get_confluence_user' vs 'find_confluence_users' and 'list_issues_by_user_role' vs 'search_issues_by_user_involvement' could cause minor confusion, but descriptions help disambiguate.
Consistent verb_noun pattern with snake_case throughout. Prefixes like add_, create_, get_, list_, search_ are uniform. Minor variation between 'get' and 'read' for different resource types but this aligns with standard conventions.
40 tools is high, covering both Confluence and Jira extensively. While each tool serves a distinct purpose, the count feels heavy for a single server, bordering on overwhelming. A more focused subset might improve usability.
Covers many operations (create, read, search, list, add comments/labels/attachments) but lacks update and delete for Jira issues, and no issue transitions. These are notable gaps that may hinder agent workflows.
Connect to Atlassian Jira, Confluence, Loom, and more to search, create, and manage your work.
Connect your team's living knowledge base — docs, data, issues, CRM — to Claude and ChatGPT.
Streamline your Attio workflows using natural language to search, create, update, and organize com…
Connect Claude, ChatGPT or any AI assistant to Zendesk tickets, CSAT, macros and Help Center.
