jira-mcp-server
Enables Claude Code CLI to interact with Jira, including fetching issue details, searching issues with JQL, retrieving comments, and getting user info, with support for multiple Jira environments.
README
Jira MCP Server
For Claude Code CLI
A Model Context Protocol (MCP) server that provides seamless Jira integration for Claude Code CLI.
Features
- Get Issue Details - Fetch complete information about any Jira issue
- Search Issues - Query issues using JQL (Jira Query Language)
- Get Comments - Retrieve all comments on an issue
- User Info - Get authenticated user details
- Multi-Environment Support - Connect to multiple Jira instances simultaneously
- Project-Specific Configuration - Different Jira environments per project via
.mcp.json
Quick Start
1. Install
cd /var/home/mrwilde/mcp/jira-mcp-server
./install.sh
2. Configure (Per-Project)
Create .mcp.json in your project root:
cd /path/to/your/project
cat > .mcp.json <<'EOF'
{
"mcpServers": {
"jira": {
"command": "python",
"args": ["-m", "jira_mcp.server"],
"env": {
"JIRA_HOST": "https://your-company.atlassian.net",
"JIRA_EMAIL": "your-email@company.com",
"JIRA_API_TOKEN": "your-api-token",
"JIRA_DEFAULT_PROJECT": "PROJ"
}
}
}
}
EOF
# Don't commit tokens!
echo ".mcp.json" >> .gitignore
3. Verify
claude mcp list
4. Use
In Claude Code CLI, simply ask:
- "Get details for issue LXP-449"
- "Search for all my open tasks"
- "Show comments on PROJ-123"
Installation & Configuration
See INSTALLATION.md for complete setup instructions including:
- System-wide installation
- Per-project configuration (
.mcp.json) - Global user configuration
- Multiple Jira environments
- CLI commands (
claude mcp add) - Security best practices
Configuration Methods
Method 1: Project .mcp.json File (Recommended)
{
"mcpServers": {
"jira": {
"command": "python",
"args": ["-m", "jira_mcp.server"],
"env": {
"JIRA_HOST": "https://company.atlassian.net",
"JIRA_EMAIL": "you@company.com",
"JIRA_API_TOKEN": "your-token"
}
}
}
}
Method 2: Claude Code CLI Command
# Add to current project
cd /path/to/project
claude mcp add --scope project jira \
--env JIRA_HOST=https://company.atlassian.net \
--env JIRA_EMAIL=you@company.com \
--env JIRA_API_TOKEN=your-token \
-- python -m jira_mcp.server
# Add globally for all projects
claude mcp add --scope user jira \
--env JIRA_HOST=https://company.atlassian.net \
--env JIRA_EMAIL=you@company.com \
--env JIRA_API_TOKEN=your-token \
-- python -m jira_mcp.server
Development
Testing
# Test connection and authentication
source venv/bin/activate
python tests/test_connection.py
# Test with specific issue
python tests/test_connection.py LXP-449
Project Structure
jira-mcp-server/
├── src/jira_mcp/
│ ├── server.py # MCP server implementation
│ └── jira_client.py # Jira API wrapper
├── tests/
│ └── test_connection.py
├── config-examples/ # Configuration templates
│ ├── project-mcp.json
│ └── multi-environment-mcp.json
├── pyproject.toml
├── requirements.txt
├── install.sh # Quick install script
└── INSTALLATION.md # Complete setup guide
MCP Tools
get_issue
Fetch detailed information about a Jira issue.
Input: issue_key (string) - e.g., "LXP-449"
Returns: Full issue details including summary, description, status, assignee, priority, dates, and project info.
search_issues
Search for issues using JQL (Jira Query Language).
Input:
jql(string) - Query string, e.g., "project = LXP AND status = 'To Do'"max_results(number, optional) - Default: 50
Returns: List of matching issues with key details.
get_issue_comments
Get all comments on a specific issue.
Input: issue_key (string)
Returns: List of comments with author, body, and timestamps.
get_current_user
Get information about the authenticated user.
Input: None
Returns: User details including username, email, and display name.
JQL Query Examples
# All issues in a project
project = LXP
# Your open issues
assignee = currentUser() AND status != Done
# Recent updates
updated >= -7d ORDER BY updated DESC
# Complex query
project = LXP AND status = "In Progress" AND assignee = currentUser()
Multiple Environments Example
{
"mcpServers": {
"jira-prod": {
"command": "python",
"args": ["-m", "jira_mcp.server"],
"env": {
"JIRA_HOST": "https://prod.atlassian.net",
"JIRA_EMAIL": "you@company.com",
"JIRA_API_TOKEN": "prod-token",
"JIRA_DEFAULT_PROJECT": "LXP"
}
},
"jira-staging": {
"command": "python",
"args": ["-m", "jira_mcp.server"],
"env": {
"JIRA_HOST": "https://staging.atlassian.net",
"JIRA_EMAIL": "you@company.com",
"JIRA_API_TOKEN": "staging-token",
"JIRA_DEFAULT_PROJECT": "LXL"
}
}
}
}
Claude Code CLI Commands
# List MCP servers
claude mcp list
# Get server details
claude mcp get jira
# Remove server
claude mcp remove --scope project jira
# Debug MCP issues
claude --mcp-debug
Getting a Jira API Token
- Visit: https://id.atlassian.com/manage-profile/security/api-tokens
- Click "Create API token"
- Name it (e.g., "MCP Server - Project X")
- Copy the token immediately
- Add it to your
.mcp.json
Security
- Never commit tokens to version control
- Add
.mcp.jsonto.gitignorein every project - Use project-specific tokens with minimal permissions
- Rotate tokens regularly (every 90 days)
- Revoke unused tokens
- Create
.mcp.json.exampletemplates for team onboarding
Requirements
- Python >= 3.10
- Claude Code CLI
- Jira Cloud account with API access
- API token for authentication
Dependencies
mcp>= 1.0.0 - Model Context Protocol SDKjira>= 3.5.0 - Python Jira librarypython-dotenv>= 1.0.0 - Environment variable managementhttpx>= 0.27.0 - HTTP client
Troubleshooting
Module Not Found
Use full Python path in .mcp.json:
{
"mcpServers": {
"jira": {
"command": "/var/home/mrwilde/mcp/jira-mcp-server/venv/bin/python",
"args": ["-m", "jira_mcp.server"],
...
}
}
}
Check MCP Status
claude mcp list
claude --mcp-debug
See INSTALLATION.md for detailed troubleshooting.
License
MIT
Contributing
Contributions welcome! Please submit issues or pull requests.
Support
- Check INSTALLATION.md for configuration help
- Review CLAUDE.md for development guidance
- Open an issue on GitHub
Changelog
0.1.0 (2025-10-28)
- Initial release for Claude Code CLI
- Core MCP server implementation
- Four essential Jira tools (get_issue, search_issues, get_issue_comments, get_current_user)
- Multi-environment support via
.mcp.json - Project-specific configuration
- Comprehensive documentation
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。