MCP Server Collection

This repository contains MCP (Model Context Protocol) servers that connect to various Azure services, exposing resources and providing tools for AI assistants.

Available MCP Servers

API Tools MCP Server

Provides powerful tools for API integration and web service interaction.

Features:

• API Integration with multiple authentication methods
• Support for REST API calls to various services
• Multiple authentication methods including Bearer token, Basic auth, and Interactive device code auth
• Comprehensive error handling with retries for transient failures
• OpenAPI schema discovery and integration

Available Tools:

• call_api - Makes authenticated API calls to a specified endpoint with various authentication options
  • Support for common HTTP methods (GET, POST, PUT, PATCH, DELETE)
  • Handles headers, query parameters, request body, and authentication configurations
• fetch_openapi_schema - Fetches and parses an OpenAPI schema from a URL with auto-discovery capabilities
• get_openapi_endpoints - Lists all available endpoints from an OpenAPI schema with their HTTP methods
• get_openapi_operations - Gets detailed information about operations for a specific endpoint path
• get_openapi_operation_details - Gets comprehensive details about a specific operation by its operationId

Azure DevOps MCP Server

Connects to Azure DevOps, exposing projects, repositories, and work items as resources.

Features:

• Connection to Azure DevOps using browser-based authentication
• Project resources (projects, repositories, work items)
• Tools for querying and working with work items
• Tools for interacting with repositories and pull requests
• Project management tools

Available Tools:

• getProjects - Lists all available projects in your Azure DevOps organization
• getRepositories - Gets repositories for a specified project
• getWorkItems - Queries work items based on filters and criteria
• createWorkItem - Creates a new work item in a project
• getPullRequests - Gets pull requests for a repository with filtering options

Kusto MCP Server

Connects to Azure Data Explorer (Kusto) database, exposing table schemas and providing data analysis tools.

Features:

• Connection to Kusto Database using service principal or managed identity authentication
• Schema resources (table schemas and sample data)
• KQL Query tools for running read-only queries
• Data analysis tools for common tasks
• Analysis prompts and templates

Available Tools:

• executeQuery - Executes a read-only KQL query against your database
• getTableInfo - Gets detailed schema and sample data for a specified table
• findTables - Finds tables that match a specified name pattern
• analyzeData - Performs various data analyses including summary statistics, time series analysis, top values analysis, outlier detection, and correlation analysis

PDF Tools MCP Server

Provides PDF processing capabilities, enabling AI assistants to extract and manipulate PDF document content.

Features:

• Text extraction from PDF documents
• Form filling capabilities for PDF forms
• PDF form element discovery and manipulation
• Support for both local and remote PDF files

Available Tools:

• extractText - Extracts text content from PDF documents with page information
• fillPdfForm - Fills in form fields in PDF documents (text fields, checkboxes, radio buttons)
• getPdfFormElements - Identifies and lists all form elements in a PDF document
• extractTables - Extracts tabular data from PDF documents
• getMetadata - Retrieves document metadata from PDF files

Setup Instructions

Prerequisites

• VS Code (latest version recommended)
• Node.js 18.0 or higher
• npm 8.0 or higher
• Access to appropriate Azure services
• Proper authentication credentials

Installation with GitHub Copilot UI

1. Ensure you have the GitHub Copilot extension installed in VS Code

   • If not, open VS Code Extensions view (Ctrl+Shift+X)
   • Search for "GitHub Copilot"
   • Click "Install"

2. Open VS Code and the GitHub Copilot Chat panel

   • Use the keyboard shortcut (Ctrl+Shift+I) or
   • Click on the Copilot Chat icon in the activity bar

3. Select "Agent Mode" in the Copilot Chat panel.

4. Click on the "Tools" icon and select Add More Tools.

5. Click Add MCP Server tool.

6. Choose Command (stdio) as the tool type.

7. Type the appropriate command to install and run your desired MCP server:

   For API Tools:

   npx @mcp-apps/api-tools-mcp-server
   

   For Azure DevOps:

   npx @mcp-apps/azure-devops-mcp-server
   

   For Kusto:

   npx @mcp-apps/kusto-mcp-server
   

   For PDF Tools:

   npx @mcp-apps/pdf-tools-mcp-server
   

8. Follow the browser authentication prompts to connect to your Azure resources.

9. Once authenticated, Copilot will be able to assist with tasks specific to the connected service.

Security Considerations

• These servers only allow operations permitted by your authentication permissions.
• Basic security measures prevent destructive operations.
• Authentication uses Azure AD for secure access.
• Consider additional security measures depending on your specific requirements.

Contributing

We welcome contributions to the MCP Server Collection! Please see CONTRIBUTING.md for guidelines and best practices on how to contribute to this project.

License

ISC