Azure DevOps Bug Tracker for GitHub Copilot Agent (MCP)
Integrates GitHub Copilot Agent with Azure DevOps to fetch, analyze, and report bug data, allowing you to query bugs, generate dashboards, and export reports via natural language.
README
Azure DevOps Bug Tracker for GitHub Copilot Agent (MCP)
A complete solution that integrates GitHub Copilot Agent with Azure DevOps to fetch, analyze, and report bug data using the Model Context Protocol (MCP).
📁 Project Structure
azure-devops-copilot-bugs/
├── .vscode/
│ └── mcp.json # MCP server config for Copilot Agent
├── .github/
│ └── copilot-instructions.md # Copilot Agent instructions
├── src/
│ ├── __init__.py
│ ├── config.py # Configuration & env management
│ ├── azure_devops_client.py # REST API client with retry
│ └── bug_report_generator.py # Multi-format report generator
├── output/ # Generated reports
├── main.py # CLI entry point
├── requirements.txt # Python dependencies
├── .env.example # Environment variable template
└── README.md # This file
🚀 Installation
Prerequisites
- Python 3.10+
- Node.js 18+ (for MCP server)
- VS Code with GitHub Copilot extension
- Azure DevOps account with PAT token
Step 1: Clone and Install
cd azure-devops-copilot-bugs
pip install -r requirements.txt
Step 2: Configure Environment Variables
# Copy the example env file
cp .env.example .env
# Edit .env with your values
Required variables:
| Variable | Description | Example |
|---|---|---|
AZURE_DEVOPS_ORG_URL |
Organization URL | https://dev.azure.com/myorg |
AZURE_DEVOPS_PAT |
Personal Access Token | xxxxxxxxxxxxxxxxxxxxx |
AZURE_DEVOPS_PROJECT |
Project name | MyProject |
Step 3: Create Azure DevOps PAT
- Go to Azure DevOps → User Settings → Personal Access Tokens
- Click "New Token"
- Set the following scopes:
- Work Items: Read
- Project and Team: Read
- Copy the generated token to your
.envfile
Step 4: Configure MCP Server (for Copilot Agent)
The .vscode/mcp.json file is pre-configured. Set environment variables in your system or VS Code settings:
Option A: System Environment Variables (Recommended for production)
[System.Environment]::SetEnvironmentVariable("AZURE_DEVOPS_ORG_URL", "https://dev.azure.com/yourorg", "User")
[System.Environment]::SetEnvironmentVariable("AZURE_DEVOPS_PAT", "your-pat-token", "User")
[System.Environment]::SetEnvironmentVariable("AZURE_DEVOPS_PROJECT", "your-project", "User")
Option B: VS Code Settings
Add to your .vscode/settings.json:
{
"github.copilot.chat.mcpServers.env": {
"AZURE_DEVOPS_ORG_URL": "https://dev.azure.com/yourorg",
"AZURE_DEVOPS_PAT": "your-pat-token",
"AZURE_DEVOPS_PROJECT": "your-project"
}
}
🖥️ CLI Usage
# Fetch active bugs
python main.py --active
# Fetch P1 and P2 bugs
python main.py --p1p2
# Fetch my bugs
python main.py --my-bugs
# Fetch current sprint bugs
python main.py --sprint
# Fetch specific sprint bugs
python main.py --sprint "MyProject\\Sprint 15"
# Show bug statistics
python main.py --stats
# List saved queries
python main.py --saved-queries
# Export all bugs to all formats
python main.py --all --export all
# Export active bugs to HTML dashboard
python main.py --active --export html
# Export P1/P2 bugs to Excel
python main.py --p1p2 --export excel
🤖 GitHub Copilot Agent Prompts
Use these prompts in VS Code Copilot Chat (Agent Mode):
Basic Queries
@workspace Show me all active bugs in Azure DevOps
@workspace List all P1 and P2 bugs with their assignees
@workspace What bugs are assigned to me?
@workspace Show bugs in the current sprint
Dashboard & Reports
@workspace Generate a daily bug dashboard with statistics
@workspace Create a summary of sprint defects grouped by priority
@workspace Show bug trends by severity for the last 3 sprints
Analysis
@workspace List all blocker bugs that are unassigned
@workspace Which team member has the most open bugs?
@workspace Show me the bug resolution rate for this sprint
@workspace Compare P1 bugs between current and previous sprint
Reporting
@workspace Export all active bugs to an Excel file with priority highlighting
@workspace Generate an HTML bug dashboard for the team standup
@workspace Create a CSV of all bugs created this week
🔧 Troubleshooting
Common Issues
| Issue | Solution |
|---|---|
401 Unauthorized |
Check PAT token is valid and has correct scopes |
404 Not Found |
Verify organization URL and project name |
Connection timeout |
Check network/VPN connectivity to Azure DevOps |
| MCP server not starting | Ensure Node.js 18+ is installed, run npx -y @anthropic/azure-devops-mcp-server manually to test |
No bugs found |
Verify WIQL query syntax and project has bug work items |
Verify PAT Token
curl -u :YOUR_PAT_TOKEN "https://dev.azure.com/yourorg/yourproject/_apis/wit/wiql?api-version=7.1" -H "Content-Type: application/json" -d "{\"query\": \"SELECT [System.Id] FROM WorkItems WHERE [System.WorkItemType] = 'Bug' AND [System.State] = 'Active'\"}"
Debug Logging
Set LOG_LEVEL=DEBUG in your .env file for verbose output:
LOG_LEVEL=DEBUG python main.py --active
MCP Server Debugging
Test the MCP server connection:
npx -y @anthropic/azure-devops-mcp-server --help
Check VS Code Output panel → "GitHub Copilot Chat" for MCP connection logs.
📊 Report Outputs
| Format | Description | Use Case |
|---|---|---|
| CSV | Comma-separated values | Data import, spreadsheets |
| Excel | Formatted .xlsx with colors | Stakeholder reports |
| HTML | Interactive dashboard | Team standup, sharing |
| JSON | Structured data | API integration, automation |
🔐 Security Notes
- Never commit
.envfiles or PAT tokens to source control - Use environment variables or Azure Key Vault for production
- PAT tokens should have minimum required scopes (Work Items: Read)
- Rotate PAT tokens regularly (recommended: every 90 days)
- The
.gitignoreexcludes sensitive files
📝 License
MIT License
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。