Jira MCP Server
An MCP server for integrating with Jira Server instances, enabling natural language interactions to create, update, search, and manage issues and comments.
README
Jira MCP Server
📚 Educational Project Notice
This repository was created as a learning exercise and educational project. While functional, please note that Atlassian provides an official MCP server for Jira integration, which may be more suitable for production use.
An MCP (Model Context Protocol) server for integrating with Jira Server instances.
Features
- Create new issues with components and labels
- Read/search existing issues
- Update issue fields, status, components, and labels
- Manage comments on issues (add, edit, delete)
- Get current user information
- JQL (Jira Query Language) support
Setup
-
Install dependencies:
npm install -
Option A: Add to Claude Code MCP configuration:
claude mcp add jira-server \ node /path/to/jira-mcp-server/server.js \ --env JIRA_BASE_URL=https://your-jira-server.com \ --env JIRA_API_TOKEN=your-api-token \ --scope userReplace:
/path/to/jira-mcp-server/with the actual path to this directoryhttps://your-jira-server.comwith your actual Jira server URL (e.g., https://jira.acme.com, https://acme-jira.example.com)your-api-tokenwith your API token
-
Option B: Add to Cursor MCP configuration:
Add this to your Cursor settings (Cursor → Settings → Features → Model Context Protocol):
{ "mcpServers": { "jira-server": { "command": "node", "args": ["/path/to/jira-mcp-server/server.js"], "env": { "JIRA_BASE_URL": "https://your-jira-server.com", "JIRA_API_TOKEN": "your-api-token" } } } }Replace the same placeholders as above. After adding this configuration:
- Restart Cursor
- Open a new chat and verify the tools are available
- If you see "No tools or prompts", check the MCP logs in Cursor's output panel
-
Option C: Run server locally for testing:
JIRA_BASE_URL=https://your-jira-server.com \ JIRA_API_TOKEN=your-api-token \ npm start
What You Can Do
This server enables your AI assistant to handle all core Jira operations through natural language:
- Create and update issues - "Create a bug about the login form" or "Update PROJ-123 to in progress and add clusters-service component"
- Search and retrieve issues - "Show me open bugs assigned to me" or "Get details for PROJ-456"
- Manage comments - "Add a comment to PROJ-123 about the fix being deployed", "Edit comment 123456", or "Delete the wrong comment"
- Get user information - "Who am I logged in as?"
Your AI assistant will automatically handle the technical details - no need to learn Jira's query language or API parameters.
Usage Examples
Once configured in Claude Code, you can use natural language to interact with Jira:
Create an issue:
"Create a new bug in project MYPROJ with the summary 'Fix login bug' and description 'Users cannot log in with special characters' with high priority"
Search issues:
"Search for all open issues in project MYPROJ" or "Find issues using JQL: project = MYPROJ AND assignee = currentUser()"
Update issue:
"Update issue MYPROJ-123 to set the status to 'In Progress', add component 'frontend', and add label 'urgent'"
Manage comments:
"Add a comment to issue MYPROJ-123 saying, 'Working on this now'" "Edit comment 123456 to say 'Updated: Still working on this'" "Delete comment 789012 - it was posted by mistake"
Get user info:
"Who am I logged in as?" or "Get my current user information"
The MCP server will automatically translate these requests into the appropriate Jira API calls using the tool parameters below:
Tool Parameters Reference:
- create_issue:
{"project": "MYPROJ", "summary": "Fix login bug", "issueType": "Bug", "priority": "High"} - search_issues:
{"jql": "project = MYPROJ AND status = 'Open'", "maxResults": 25} - update_issue:
{"issueKey": "MYPROJ-123", "summary": "Updated summary", "status": "In Progress", "components": ["frontend"], "labels": ["urgent"]} - add_comment:
{"issueKey": "MYPROJ-123", "body": "Working on this now"} - edit_comment:
{"issueKey": "MYPROJ-123", "commentId": "123456", "body": "Updated comment"} - delete_comment:
{"issueKey": "MYPROJ-123", "commentId": "789012"} - get_comments:
{"issueKey": "MYPROJ-123"} - get_current_user: No parameters required
Security
Important: This server requires API credentials to access your Jira instance. Please follow these security best practices:
- Never commit API tokens to version control - Always use environment variables
- Use least-privilege access - Create API tokens with minimal required permissions
- Rotate tokens regularly - Generate new API tokens periodically
- Secure your environment - Ensure your
.envfiles and environment variables are properly secured - Review token scope - Only grant access to projects and operations you need
Requirements
- Node.js: Version 18.0.0 or higher
- Jira Server: Compatible with Jira Server instances (not Jira Cloud)
- API Token: Valid Jira API token for authentication
Troubleshooting
Connection Issues
- ECONNREFUSED: Check that your
JIRA_BASE_URLis correct and accessible - SSL/TLS errors: Verify your Jira server's SSL certificate configuration
- Timeout errors: Check network connectivity to your Jira server
Authentication Issues
- 401 Unauthorized: Verify your
JIRA_API_TOKENis valid and hasn't expired - 403 Forbidden: Check that your API token has sufficient permissions for the requested operation
- Invalid credentials: Ensure the API token is correctly copied (no extra spaces or characters)
Common Setup Issues
- Module not found: Run
npm installto install dependencies - Permission denied: Check file permissions on the server.js file
- Environment variables not loaded: Ensure environment variables are properly set in your shell or MCP configuration
Cursor-Specific Issues
- "No tools or prompts" error:
- Verify the file path in your MCP configuration is absolute and correct
- Check Cursor's output panel for MCP server errors
- Ensure Node.js is in your PATH and accessible from Cursor
- Try restarting Cursor after configuration changes
- MCP server not starting:
- Check that all file paths use forward slashes (even on Windows)
- Verify environment variables are properly set in the configuration
- Test the server manually:
node /path/to/server.jsto see any startup errors
Getting Help
- Check the Jira Server logs for detailed error messages
- Verify API token permissions in your Jira admin panel
- For Cursor users: Check the MCP output panel in Cursor for detailed error logs
- Check the detailed documentation for advanced troubleshooting
Contributing
Contributions are welcome! Please see CONTRIBUTING.md for detailed guidelines on how to contribute to this project.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Notes
- This server uses Jira REST API v2
- Authentication uses an API token
- Supports Jira Server (not Cloud-specific features)
- All responses are returned as JSON text
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。