MCP Atlassian
MCP Atlassian
适用于 Atlassian 产品(Confluence 和 Jira)的模型上下文协议 (MCP) 服务器。此集成支持 Confluence & Jira Cloud 以及服务器/数据中心部署。
示例用法
要求你的人工智能助手:
📝 自动 Jira 更新- “从我们的会议记录更新 Jira”
🔍 人工智能驱动的 Confluence 搜索- “在 Confluence 中查找我们的 OKR 指南并进行总结”
🐛 智能 Jira 问题过滤- “显示上周 PROJ 项目中的紧急错误”
📄 内容创建和管理- “为 XYZ 功能创建技术设计文档”
功能演示
https://github.com/user-attachments/assets/35303504-14c6-4ae4-913b-7c25ea511c3e
https://github.com/user-attachments/assets/7fe9c488-ad0c-4876-9b54-120b666bb785
兼容性
产品 | 部署类型 | 支持状态 |
合流 | 云 | ✅ 完全支持 |
合流 | 服务器/数据中心 | ✅ 支持(版本 6.0+) |
吉拉 | 云 | ✅ 完全支持 |
吉拉 | 服务器/数据中心 | ✅ 支持(版本 8.14+) |
Related MCP server: MCP Atlassian
快速入门指南
🔐 1. 身份验证设置
MCP Atlassian 支持三种身份验证方法:
A. API 令牌认证(云)
前往https://id.atlassian.com/manage-profile/security/api-tokens
点击创建 API 令牌,并命名
立即复制令牌
B.个人访问令牌(服务器/数据中心)
前往您的个人资料(头像)→个人资料→个人访问令牌
单击创建令牌,命名,设置有效期
立即复制令牌
C.OAuth 2.0身份验证(云)
创建“OAuth 2.0 (3LO) 集成”应用
为 Jira/Confluence 配置权限(范围)
设置回调 URL (例如
http://localhost:8080/callback)运行安装向导:
docker run --rm -i \ -p 8080:8080 \ -v "${HOME}/.mcp-atlassian:/home/app/.mcp-atlassian" \ ghcr.io/sooperset/mcp-atlassian:latest --oauth-setup -v按照提示输入
Client ID、Secret、URI和Scope完成浏览器授权
将获取的凭证添加到
.env或 IDE 配置中:ATLASSIAN_OAUTH_CLOUD_ID(来自向导)ATLASSIAN_OAUTH_CLIENT_IDATLASSIAN_OAUTH_CLIENT_SECRETATLASSIAN_OAUTH_REDIRECT_URIATLASSIAN_OAUTH_SCOPE
[!IMPORTANT] 在持久身份验证范围内包含
offline_access(例如read:jira-work write:jira-work offline_access)
📦 2. 安装
MCP Atlassian 以 Docker 镜像的形式分发。这是运行服务器的推荐方式,尤其适用于 IDE 集成。请确保已安装 Docker。
# Pull Pre-built Image
docker pull ghcr.io/sooperset/mcp-atlassian:latest🛠️ IDE 集成
MCP Atlassian 旨在通过 IDE 集成与 AI 助手一起使用。
[!TIP]对于 Claude Desktop :直接找到并编辑配置文件:
Windows :
%APPDATA%\Claude\claude_desktop_config.jsonmacOS :
~/Library/Application Support/Claude/claude_desktop_config.jsonLinux :
~/.config/Claude/claude_desktop_config.json对于光标:打开设置 → MCP → + 添加新的全局 MCP 服务器
⚙️ 配置方法
配置 Docker 容器主要有两种方法:
直接传递变量(如下例所示)
使用带有
--env-file标志的环境文件(显示在可折叠部分)
常见的环境变量包括:
CONFLUENCE_SPACES_FILTER:按空格键过滤(例如“DEV、TEAM、DOC”)
JIRA_PROJECTS_FILTER:按项目键过滤(例如“PROJ、DEV、SUPPORT”)
READ_ONLY_MODE:设置为“true”以禁用写入操作
MCP_VERBOSE:设置为“true”以获得更详细的日志记录
ENABLED_TOOLS:要启用的工具名称的逗号分隔列表(例如“confluence_search,jira_get_issue”)请参阅.env.example文件以了解所有可用选项。
📝 配置示例
方法 1(直接传递变量):
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "CONFLUENCE_URL",
"-e", "CONFLUENCE_USERNAME",
"-e", "CONFLUENCE_API_TOKEN",
"-e", "JIRA_URL",
"-e", "JIRA_USERNAME",
"-e", "JIRA_API_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
"CONFLUENCE_USERNAME": "your.email@company.com",
"CONFLUENCE_API_TOKEN": "your_confluence_api_token",
"JIRA_URL": "https://your-company.atlassian.net",
"JIRA_USERNAME": "your.email@company.com",
"JIRA_API_TOKEN": "your_jira_api_token"
}
}
}
}{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--env-file",
"/path/to/your/mcp-atlassian.env",
"ghcr.io/sooperset/mcp-atlassian:latest"
]
}
}
}对于服务器/数据中心部署,使用直接变量传递:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "CONFLUENCE_URL",
"-e", "CONFLUENCE_PERSONAL_TOKEN",
"-e", "CONFLUENCE_SSL_VERIFY",
"-e", "JIRA_URL",
"-e", "JIRA_PERSONAL_TOKEN",
"-e", "JIRA_SSL_VERIFY",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"CONFLUENCE_URL": "https://confluence.your-company.com",
"CONFLUENCE_PERSONAL_TOKEN": "your_confluence_pat",
"CONFLUENCE_SSL_VERIFY": "false",
"JIRA_URL": "https://jira.your-company.com",
"JIRA_PERSONAL_TOKEN": "your_jira_pat",
"JIRA_SSL_VERIFY": "false"
}
}
}
}[!NOTE] 仅当您拥有自签名证书时,才将
CONFLUENCE_SSL_VERIFY和JIRA_SSL_VERIFY设置为“false”。
本示例展示了如何在您的 IDE(例如 Cursor 或 Claude Desktop)中配置mcp-atlassian并在 Atlassian Cloud 上使用 OAuth 2.0。请确保您已完成OAuth 设置向导。
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-v", "<path_to_your_home>/.mcp-atlassian:/home/app/.mcp-atlassian",
"-e", "JIRA_URL",
"-e", "CONFLUENCE_URL",
"-e", "ATLASSIAN_OAUTH_CLIENT_ID",
"-e", "ATLASSIAN_OAUTH_CLIENT_SECRET",
"-e", "ATLASSIAN_OAUTH_REDIRECT_URI",
"-e", "ATLASSIAN_OAUTH_SCOPE",
"-e", "ATLASSIAN_OAUTH_CLOUD_ID",
"ghcr.io/sooperset/mcp-atlassian:latest",
],
"env": {
"JIRA_URL": "https://your-company.atlassian.net",
"CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
"ATLASSIAN_OAUTH_CLIENT_ID": "YOUR_OAUTH_APP_CLIENT_ID",
"ATLASSIAN_OAUTH_CLIENT_SECRET": "YOUR_OAUTH_APP_CLIENT_SECRET",
"ATLASSIAN_OAUTH_REDIRECT_URI": "http://localhost:8080/callback",
"ATLASSIAN_OAUTH_SCOPE": "read:jira-work write:jira-work read:confluence-content.all write:confluence-content offline_access",
"ATLASSIAN_OAUTH_CLOUD_ID": "YOUR_CLOUD_ID_FROM_SETUP_WIZARD"
}
}
}
}[!笔记]
ATLASSIAN_OAUTH_CLOUD_ID是从--oauth-setup向导输出中获得的。其他
ATLASSIAN_OAUTH_*变量是您在 Atlassian 开发者控制台中为您的 OAuth 应用配置的变量(并用作安装向导的输入)。您的云实例的
JIRA_URL和CONFLUENCE_URL仍然是必需的。
MCP Atlassian 支持通过标准 HTTP/HTTPS/SOCKS 代理路由 API 请求。使用环境变量进行配置:
支持标准
HTTP_PROXY、HTTPS_PROXY、NO_PROXY、SOCKS_PROXY。可以使用特定于服务的覆盖(例如,
JIRA_HTTPS_PROXY,CONFLUENCE_NO_PROXY)。特定于服务的变量会覆盖该服务的全局变量。
将相关的代理变量添加到 MCP 配置的args (使用-e )和env部分:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "... existing Confluence/Jira vars",
"-e", "HTTP_PROXY",
"-e", "HTTPS_PROXY",
"-e", "NO_PROXY",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"... existing Confluence/Jira vars": "...",
"HTTP_PROXY": "http://proxy.internal:8080",
"HTTPS_PROXY": "http://proxy.internal:8080",
"NO_PROXY": "localhost,.your-company.com"
}
}
}
}代理 URL 中的凭证在日志中会被屏蔽。如果您设置了NO_PROXY ,则对匹配主机的请求将遵循该设置。
仅适用于 Confluence Cloud:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "CONFLUENCE_URL",
"-e", "CONFLUENCE_USERNAME",
"-e", "CONFLUENCE_API_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
"CONFLUENCE_USERNAME": "your.email@company.com",
"CONFLUENCE_API_TOKEN": "your_api_token"
}
}
}
}对于 Confluence Server/DC,使用:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "CONFLUENCE_URL",
"-e", "CONFLUENCE_PERSONAL_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"CONFLUENCE_URL": "https://confluence.your-company.com",
"CONFLUENCE_PERSONAL_TOKEN": "your_personal_token"
}
}
}
}仅适用于 Jira Cloud:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "JIRA_URL",
"-e", "JIRA_USERNAME",
"-e", "JIRA_API_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"JIRA_URL": "https://your-company.atlassian.net",
"JIRA_USERNAME": "your.email@company.com",
"JIRA_API_TOKEN": "your_api_token"
}
}
}
}对于 Jira Server/DC,使用:
{
"mcpServers": {
"mcp-atlassian": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e", "JIRA_URL",
"-e", "JIRA_PERSONAL_TOKEN",
"ghcr.io/sooperset/mcp-atlassian:latest"
],
"env": {
"JIRA_URL": "https://jira.your-company.com",
"JIRA_PERSONAL_TOKEN": "your_personal_token"
}
}
}
}👥 HTTP 传输配置
除了使用stdio之外,您还可以使用以下任一方式将服务器作为持久 HTTP 服务运行:
/sse端点的sse(服务器发送事件)传输/mcp端点上的streamable-http传输
两种传输类型都支持单用户和多用户身份验证:
身份验证选项:
单用户:使用通过环境变量配置的服务器级身份验证
多用户:每个用户提供自己的身份验证:
云:OAuth 2.0 Bearer 令牌
服务器/数据中心:个人访问令牌(PAT)
使用您选择的传输方式启动服务器:
# For SSE transport docker run --rm -p 9000:9000 \ --env-file /path/to/your/.env \ ghcr.io/sooperset/mcp-atlassian:latest \ --transport sse --port 9000 -vv # OR for streamable-http transport docker run --rm -p 9000:9000 \ --env-file /path/to/your/.env \ ghcr.io/sooperset/mcp-atlassian:latest \ --transport streamable-http --port 9000 -vv配置您的 IDE(单用户示例):
SSE 传输示例:
{ "mcpServers": { "mcp-atlassian-http": { "url": "http://localhost:9000/sse" } } }Streamable-HTTP 传输示例:
{ "mcpServers": { "mcp-atlassian-service": { "url": "http://localhost:9000/mcp" } } }
以下是使用流式 HTTP 传输设置多用户身份验证的完整示例:
首先,运行 OAuth 设置向导来配置服务器的 OAuth 凭据:
docker run --rm -i \ -p 8080:8080 \ -v "${HOME}/.mcp-atlassian:/home/app/.mcp-atlassian" \ ghcr.io/sooperset/mcp-atlassian:latest --oauth-setup -v使用流式 HTTP 传输启动服务器:
docker run --rm -p 9000:9000 \ --env-file /path/to/your/.env \ ghcr.io/sooperset/mcp-atlassian:latest \ --transport streamable-http --port 9000 -vv配置 IDE 的 MCP 设置:
为您的 Atlassian 部署选择适当的授权方法:
**云(OAuth 2.0):**如果您的组织在 Atlassian Cloud 上,并且您为每个用户都有一个 OAuth 访问令牌,请使用此功能。
**服务器/数据中心(PAT):**如果您在 Atlassian 服务器或数据中心并且每个用户都有个人访问令牌(PAT),请使用此功能。
云(OAuth 2.0)示例:
{
"mcpServers": {
"mcp-atlassian-service": {
"url": "http://localhost:9000/mcp",
"headers": {
"Authorization": "Bearer <USER_OAUTH_ACCESS_TOKEN>"
}
}
}
}服务器/数据中心 (PAT) 示例:
{
"mcpServers": {
"mcp-atlassian-service": {
"url": "http://localhost:9000/mcp",
"headers": {
"Authorization": "Token <USER_PERSONAL_ACCESS_TOKEN>"
}
}
}
}.env中必需的环境变量:JIRA_URL=https://your-company.atlassian.net CONFLUENCE_URL=https://your-company.atlassian.net/wiki ATLASSIAN_OAUTH_CLIENT_ID=your_oauth_app_client_id ATLASSIAN_OAUTH_CLIENT_SECRET=your_oauth_app_client_secret ATLASSIAN_OAUTH_REDIRECT_URI=http://localhost:8080/callback ATLASSIAN_OAUTH_SCOPE=read:jira-work write:jira-work read:confluence-content.all write:confluence-content offline_access ATLASSIAN_OAUTH_CLOUD_ID=your_cloud_id_from_setup_wizard
[!笔记]
服务器应该配置自己的备用身份验证(例如,通过 API 令牌、PAT 的环境变量,或使用 --oauth-setup 设置自己的 OAuth)。如果请求不包含用户特定的身份验证,则使用此选项。
OAuth :每个用户都需要从您的 Atlassian OAuth 应用获取自己的 OAuth 访问令牌。
PAT :每个用户提供自己的个人访问令牌。
如果用户提供令牌,服务器将使用用户令牌进行 API 调用;如果用户未提供令牌,则返回服务器身份验证
用户令牌应具有适合其所需操作的范围
工具
关键工具
Jira 工具
jira_get_issue:获取特定问题的详细信息jira_search:使用 JQL 搜索问题jira_create_issue:创建新问题jira_update_issue:更新现有问题jira_transition_issue:将问题转换为新状态jira_add_comment:向问题添加评论
Confluence 工具
confluence_search:使用 CQL 搜索 Confluence 内容confluence_get_page:获取特定页面的内容confluence_create_page:创建新页面confluence_update_page:更新现有页面
手术 | Jira 工具 | Confluence 工具 |
读 |
|
|
|
| |
|
| |
|
| |
|
| |
| ||
| ||
| ||
| ||
| ||
| ||
| ||
| ||
写 |
|
|
|
| |
|
| |
|
| |
|
| |
| ||
| ||
| ||
| ||
| ||
| ||
|
*该工具仅在 Jira Cloud 上可用
工具过滤和访问控制
服务器提供了两种控制工具访问的方式:
工具过滤:使用
--enabled-tools标志或ENABLED_TOOLS环境变量来指定哪些工具可用:# Via environment variable ENABLED_TOOLS="confluence_search,jira_get_issue,jira_search" # Or via command line flag docker run ... --enabled-tools "confluence_search,jira_get_issue,jira_search" ...读/写控制:工具分为读操作和写操作。启用
READ_ONLY_MODE后,无论ENABLED_TOOLS设置如何,都只能进行读操作。
故障排除和调试
常见问题
身份验证失败:
对于云:检查您的 API 令牌(而不是您的帐户密码)
对于服务器/数据中心:验证您的个人访问令牌是否有效且未过期
对于较旧的 Confluence 服务器:某些旧版本需要使用
CONFLUENCE_USERNAME和CONFLUENCE_API_TOKEN进行基本身份验证(其中 token 是您的密码)
SSL 证书问题:如果使用服务器/数据中心并遇到 SSL 错误,请设置
CONFLUENCE_SSL_VERIFY=false或JIRA_SSL_VERIFY=false权限错误:确保您的 Atlassian 帐户具有足够的权限来访问空间/项目
调试工具
# Using MCP Inspector for testing
npx @modelcontextprotocol/inspector uvx mcp-atlassian ...
# For local development version
npx @modelcontextprotocol/inspector uv --directory /path/to/your/mcp-atlassian run mcp-atlassian ...
# View logs
# macOS
tail -n 20 -f ~/Library/Logs/Claude/mcp*.log
# Windows
type %APPDATA%\Claude\logs\mcp*.log | more安全
切勿共享 API 令牌
确保 .env 文件的安全性和私密性
请参阅SECURITY.md了解最佳实践
贡献
欢迎为 MCP Atlassian 做出贡献!如果您想贡献:
查看我们的CONTRIBUTING.md指南,了解详细的开发设置说明。
进行更改并提交拉取请求。
我们使用预提交钩子来保证代码质量,并遵循语义版本控制来发布。
执照
采用 MIT 许可 - 请参阅LICENSE文件。本产品非 Atlassian 官方产品。
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
A Model Context Protocol (MCP) server for Selise Blocks Cloud integration
Model Context Protocol server for the Apideck Unified API. Connect any MCP-compatible agent framework to 100+ accounting systems, HRIS platforms, file storage providers, and more through one integration. More information https://www.apideck.com/mcp-server
Model Context Protocol server for Studex tools, notifications, and profile integrations
MCP server for AI access to SmartBear tools, including BugSnag, Reflect, Swagger, PactFlow, QTM4J.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceModel Context Protocol server that integrates with Atlassian Confluence and Jira, enabling AI assistants to search, create, and update content in these platforms through natural language interactions.1MIT
- -licenseNot gradedqualityNot gradedmaintenanceA Model Context Protocol server that enables interaction with Atlassian products (Confluence and Jira), supporting both Cloud and Server/Data Center deployments for searching, creating, and managing content through natural language.-
- AlicenseBqualityCmaintenanceA Model Context Protocol server that integrates with Atlassian's Jira and Confluence, enabling AI assistants to interact with these tools directly through features like issue management, page creation, and content search.131MIT
- AlicenseNot gradedqualityDmaintenanceA Model Context Protocol server for Atlassian Jira and Confluence that supports both Cloud and On-Prem/Data Center deployments. It enables AI assistants to search, create, and manage issues and pages using secure authentication methods like PAT and OAuth.5MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/sooperset/mcp-atlassian'
If you have feedback or need assistance with the MCP directory API, please join our Discord server