Atlassian MCP Server
A read-only MCP server that enables querying and searching Atlassian Confluence pages and Jira issues through their REST APIs. Supports retrieving content by ID or URL, searching using CQL/JQL, and listing spaces and projects.
README
Atlassian MCP Server
A read-only Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira).
Installation
pip install -e .
Configuration
Set your Atlassian credentials as environment variables:
export ATLASSIAN_DOMAIN=your-company.atlassian.net
export ATLASSIAN_EMAIL=your-email@company.com
# Separate tokens for each service (recommended)
export ATLASSIAN_CONFLUENCE_TOKEN=your-confluence-token
export ATLASSIAN_JIRA_TOKEN=your-jira-token
# OR for backward compatibility, single token (if it has access to both)
export ATLASSIAN_API_TOKEN=your-api-token
Creating API Tokens
- Go to: https://id.atlassian.com/manage-profile/security/api-tokens
- Click "Create API token"
For Confluence:
- When creating a Confluence token, select these scopes:
read:content:confluence- View pages, blog posts, comments, attachmentsread:content-details:confluence- View content metadata and properties
For Jira:
- When creating a Jira token, you'll see different scope options. Look for:
- Scopes that allow reading issues and projects
- Search capabilities for JQL queries
- Basic browse permissions
Note: The exact scope names may vary in the Atlassian interface. Choose the minimal read-only scopes that allow viewing issues, projects, and searching. If you're unsure, you can start with broader read permissions and restrict them later.
Usage
Start the MCP server:
atlassian-mcp
Available Tools
Confluence
- confluence_get_page - Get a page by ID
- confluence_get_page_by_url - Get a page by URL
- confluence_search_pages - Search for pages
- confluence_list_spaces - List all spaces
Jira
- jira_get_issue - Get an issue by key (e.g., PROJ-123)
- jira_get_issue_by_url - Get an issue by URL
- jira_search_issues - Search issues using JQL
- jira_list_projects - List all projects
Supported URLs
Confluence Pages
https://domain.atlassian.net/wiki/spaces/SPACE/pages/123456/Page+Titlehttps://domain.atlassian.net/wiki/display/SPACE/Page+Title?pageId=123456https://domain.atlassian.net/wiki/pages/viewpage.action?pageId=123456
Jira Issues
https://domain.atlassian.net/browse/PROJ-123https://domain.atlassian.net/jira/software/projects/PROJ/boards/1?selectedIssue=PROJ-123
MCP Client Configuration
{
"mcpServers": {
"atlassian": {
"command": "python",
"args": ["-m", "atlassian_mcp.server"],
"cwd": "/path/to/your/atlassian_mcp",
"env": {
"ATLASSIAN_DOMAIN": "your-company.atlassian.net",
"ATLASSIAN_EMAIL": "your-email@company.com",
"ATLASSIAN_CONFLUENCE_TOKEN": "your-confluence-token",
"ATLASSIAN_JIRA_TOKEN": "your-jira-token"
}
}
}
}
API Endpoints Used
This server uses the following Atlassian REST API endpoints:
Confluence REST API:
GET /wiki/rest/api/content/{id}- Get page contentGET /wiki/rest/api/content/search- Search pages using CQLGET /wiki/rest/api/space- List spaces
Jira REST API v3:
GET /rest/api/3/issue/{issueIdOrKey}- Get issue detailsGET /rest/api/3/search- Search issues using JQLGET /rest/api/3/project- List projects
All endpoints are read-only and require basic authentication with your email and API token.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。