feat: added mcp server, and semantic kernal setup

.gitignore

@@ -6,6 +6,7 @@ __pycache__/
 # C extensions
 *.so
 uv.lock
+archieve
 
 # Distribution / packaging
 .Python

README.md

@@ -1 +1,56 @@
-# azure-agent-ops-copilot
+# Azure Ops Copilot
+
+An agentic AI application that acts as a DevOps assistant for Azure. It uses the Azure Agent Framework, Semantic Kernel, and Model Context Protocol (MCP) to analyze alerts, inspect configurations, and suggest fixes.
+
+## Features
+
+- **Log Analyzer**: Reads Azure Monitor alerts from synthetic logs.
+- **Config Reader**: Inspects resource configurations (ARM/Bicep).
+- **Fix Generator**: Suggests remediation steps (CLI/Bicep snippets).
+- **Agentic Interface**: Chat with the copilot via a web UI.
+
+## Architecture
+
+- **Backend**: FastAPI
+- **Frontend**: Gradio
+- **AI Engine**: Semantic Kernel + Azure OpenAI
+- **Tools**: MCP Server (FastMCP)
+
+## Setup
+
+1.  **Install Dependencies**:
+    ```bash
+    uv sync
+    ```
+
+2.  **Configure Environment**:
+    Copy `.env.template` to `.env` (or create it) and add your Azure OpenAI keys.
+    ```bash
+    AZURE_OPENAI_ENDPOINT=...
+    AZURE_OPENAI_API_KEY=...
+    AZURE_OPENAI_DEPLOYMENT_NAME=...
+    ```
+
+3.  **Generate Synthetic Data**:
+    ```bash
+    uv run python scripts/generate_logs.py
+    uv run python scripts/generate_configs.py
+    ```
+
+4.  **Run the Application**:
+    Start the API:
+    ```bash
+    uv run python src/api/main.py
+    ```
+
+    Start the UI (in a separate terminal):
+    ```bash
+    uv run python src/ui/app.py
+    ```
+
+## Usage
+
+Open the Gradio UI (usually at `http://localhost:7860`) and ask questions like:
+- "Analyze alert alert-001"
+- "Check config for vm-01"
+- "Suggest a fix for high CPU on vm-01"

data/configs.json

@@ -0,0 +1,43 @@
+[
+  {
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "type": "Microsoft.Compute/virtualMachines",
+    "location": "eastus",
+    "properties": {
+      "hardwareProfile": {
+        "vmSize": "Standard_D2s_v3"
+      },
+      "storageProfile": {
+        "osDisk": {
+          "osType": "Linux",
+          "diskSizeGB": 30
+        }
+      }
+    },
+    "compliance_status": "Compliant"
+  },
+  {
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+    "type": "Microsoft.Sql/servers/databases",
+    "location": "eastus",
+    "properties": {
+      "sku": {
+        "name": "Standard",
+        "tier": "Standard",
+        "capacity": 10
+      },
+      "maxSizeBytes": 2147483648
+    },
+    "compliance_status": "NonCompliant"
+  },
+  {
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/sites/app-01",
+    "type": "Microsoft.Web/sites",
+    "location": "eastus",
+    "properties": {
+      "serverFarmId": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/serverfarms/plan-01",
+      "httpsOnly": false
+    },
+    "compliance_status": "NonCompliant"
+  }
+]

data/logs.json

@@ -0,0 +1,122 @@
+[
+  {
+    "id": "alert-001",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+    "description": "High CPU usage detected (95%)",
+    "created_at": "2025-11-24T07:44:05.852238",
+    "status": "New",
+    "properties": {
+      "metric_value": 100,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-002",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "description": "High CPU usage detected (95%)",
+    "created_at": "2025-11-24T07:40:05.852399",
+    "status": "New",
+    "properties": {
+      "metric_value": 86,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-003",
+    "severity": "Warning",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/sites/app-01",
+    "description": "SQL Database DTU usage high (90%)",
+    "created_at": "2025-11-24T07:43:05.852404",
+    "status": "New",
+    "properties": {
+      "metric_value": 96,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-004",
+    "severity": "Warning",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "description": "App Service response time high (>2s)",
+    "created_at": "2025-11-24T07:42:05.852407",
+    "status": "New",
+    "properties": {
+      "metric_value": 100,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-005",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+    "description": "Disk space low (5% remaining)",
+    "created_at": "2025-11-24T07:19:05.852410",
+    "status": "New",
+    "properties": {
+      "metric_value": 90,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-006",
+    "severity": "Warning",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+    "description": "SQL Database DTU usage high (90%)",
+    "created_at": "2025-11-24T07:36:05.852412",
+    "status": "New",
+    "properties": {
+      "metric_value": 97,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-007",
+    "severity": "Warning",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "description": "App Service response time high (>2s)",
+    "created_at": "2025-11-24T06:50:05.852414",
+    "status": "New",
+    "properties": {
+      "metric_value": 91,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-008",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "description": "High CPU usage detected (95%)",
+    "created_at": "2025-11-24T07:04:05.852417",
+    "status": "New",
+    "properties": {
+      "metric_value": 89,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-009",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/sites/app-01",
+    "description": "High CPU usage detected (95%)",
+    "created_at": "2025-11-24T07:00:05.852419",
+    "status": "New",
+    "properties": {
+      "metric_value": 99,
+      "threshold": 80
+    }
+  },
+  {
+    "id": "alert-010",
+    "severity": "Critical",
+    "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "description": "High CPU usage detected (95%)",
+    "created_at": "2025-11-24T07:03:05.852421",
+    "status": "New",
+    "properties": {
+      "metric_value": 93,
+      "threshold": 80
+    }
+  }
+]

data/templates/sql_scale.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+# Scale SQL Database
+RESOURCE_GROUP=$1
+SERVER_NAME=$2
+DB_NAME=$3
+NEW_SKU=$4
+
+az sql db update \
+    --resource-group $RESOURCE_GROUP \
+    --server $SERVER_NAME \
+    --name $DB_NAME \
+    --service-objective $NEW_SKU

data/templates/vm_resize.bicep

@@ -0,0 +1,17 @@
+param vmName string
+param location string = resourceGroup().location
+param vmSize string = 'Standard_D4s_v3'
+
+resource vm 'Microsoft.Compute/virtualMachines@2021-03-01' existing = {
+  name: vmName
+}
+
+resource vmUpdate 'Microsoft.Compute/virtualMachines@2021-03-01' = {
+  name: vmName
+  location: location
+  properties: {
+    hardwareProfile: {
+      vmSize: vmSize
+    }
+  }
+}

pyproject.toml

@@ -9,6 +9,8 @@ dependencies = []
 [dependency-groups]
 backend = [
     "agent-framework>=1.0.0b251120",
+    "azure-ai-inference>=1.0.0b9",
+    "fastapi[standard]>=0.121.3",
     "semantic-kernel>=1.38.0",
 ]
 frontend = [

scripts/generate_configs.py

@@ -0,0 +1,52 @@
+import json
+from pathlib import Path
+
+# Ensure data directory exists
+DATA_DIR = Path(__file__).parent.parent / "data"
+DATA_DIR.mkdir(exist_ok=True)
+
+CONFIGS_FILE = DATA_DIR / "configs.json"
+
+CONFIGS = [
+    {
+        "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+        "type": "Microsoft.Compute/virtualMachines",
+        "location": "eastus",
+        "properties": {
+            "hardwareProfile": {"vmSize": "Standard_D2s_v3"},
+            "storageProfile": {"osDisk": {"osType": "Linux", "diskSizeGB": 30}},
+        },
+        "compliance_status": "Compliant",
+    },
+    {
+        "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+        "type": "Microsoft.Sql/servers/databases",
+        "location": "eastus",
+        "properties": {
+            "sku": {"name": "Standard", "tier": "Standard", "capacity": 10},
+            "maxSizeBytes": 2147483648,
+        },
+        "compliance_status": "NonCompliant",
+    },
+    {
+        "resource_id": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/sites/app-01",
+        "type": "Microsoft.Web/sites",
+        "location": "eastus",
+        "properties": {
+            "serverFarmId": "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/serverfarms/plan-01",
+            "httpsOnly": False,
+        },
+        "compliance_status": "NonCompliant",
+    },
+]
+
+
+def generate_configs():
+    with open(CONFIGS_FILE, "w") as f:
+        json.dump(CONFIGS, f, indent=2)
+
+    print(f"Generated {len(CONFIGS)} configs in {CONFIGS_FILE}")
+
+
+if __name__ == "__main__":
+    generate_configs()

scripts/generate_logs.py

@@ -0,0 +1,58 @@
+import json
+import random
+from datetime import datetime, timedelta
+from pathlib import Path
+from typing import Any, Dict, List
+
+# Ensure data directory exists
+DATA_DIR = Path(__file__).parent.parent / "data"
+DATA_DIR.mkdir(exist_ok=True)
+
+LOGS_FILE = DATA_DIR / "logs.json"
+
+RESOURCE_IDS = [
+    "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01",
+    "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Sql/servers/sql-01/databases/db-01",
+    "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Web/sites/app-01",
+]
+
+ALERT_DESCRIPTIONS = [
+    "High CPU usage detected (95%)",
+    "Disk space low (5% remaining)",
+    "SQL Database DTU usage high (90%)",
+    "App Service response time high (>2s)",
+]
+
+
+def generate_logs(count: int = 10):
+    logs: List[Dict[str, Any]] = []
+    for i in range(count):
+        resource_id = random.choice(RESOURCE_IDS)
+        description = random.choice(ALERT_DESCRIPTIONS)
+
+        # Determine severity based on description
+        severity = "Warning"
+        if "95%" in description or "5%" in description:
+            severity = "Critical"
+
+        log = {
+            "id": f"alert-{i + 1:03d}",
+            "severity": severity,
+            "resource_id": resource_id,
+            "description": description,
+            "created_at": (
+                datetime.now() - timedelta(minutes=random.randint(1, 60))
+            ).isoformat(),
+            "status": "New",
+            "properties": {"metric_value": random.randint(80, 100), "threshold": 80},
+        }
+        logs.append(log)
+
+    with open(LOGS_FILE, "w") as f:
+        json.dump(logs, f, indent=2)
+
+    print(f"Generated {count} logs in {LOGS_FILE}")
+
+
+if __name__ == "__main__":
+    generate_logs()

src/backend/agent/azure_agent.py

@@ -0,0 +1,93 @@
+import os
+from typing import Annotated
+
+from dotenv import load_dotenv
+from semantic_kernel import Kernel
+from semantic_kernel.agents import ChatCompletionAgent
+from semantic_kernel.connectors.ai.function_choice_behavior import (
+    FunctionChoiceBehavior,
+)
+from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion
+from semantic_kernel.functions import kernel_function
+from services.mcp_client import MCPClient
+
+load_dotenv()
+mcp_client = MCPClient()
+
+
+class AzureOpsPlugin:
+    """Plugin wrapping our MCP tools for Semantic Kernel."""
+
+    @kernel_function(
+        name="analyze_alert", description="Analyze an Azure Monitor alert by ID."
+    )
+    def analyze_alert_wrapper(
+        self, alert_id: Annotated[str, "The ID of the alert to analyze"]
+    ) -> str:
+        return mcp_client.execute("analyze_alert", alert_id=alert_id)
+
+    @kernel_function(
+        name="get_resource_config",
+        description="Get the configuration of an Azure resource.",
+    )
+    def get_resource_config_wrapper(
+        self, resource_id: Annotated[str, "The ID of the resource"]
+    ) -> str:
+        return mcp_client.execute("get_resource_config", resource_id=resource_id)
+
+    @kernel_function(
+        name="generate_fix",
+        description="Generate a fix for a specific issue type and resource type.",
+    )
+    def generate_fix_wrapper(
+        self,
+        issue_type: Annotated[str, "The type of issue (e.g., 'High CPU')"],
+        resource_type: Annotated[str, "The type of resource"],
+    ) -> str:
+        return mcp_client.execute(
+            "generate_fix", issue_type=issue_type, resource_type=resource_type
+        )
+
+
+async def get_agent() -> Kernel:
+    kernel = Kernel()
+
+    service_id = "default"
+    try:
+        kernel.add_service(
+            AzureChatCompletion(
+                service_id=service_id,
+                deployment_name=os.getenv("AZURE_OPENAI_DEPLOYMENT_NAME", "gpt-4"),
+                endpoint=os.getenv("AZURE_OPENAI_ENDPOINT", ""),
+                api_key=os.getenv("AZURE_OPENAI_API_KEY", ""),
+            )
+        )
+    except Exception as e:
+        print(f"Warning: Could not configure Azure OpenAI: {e}")
+
+    kernel.add_plugin(AzureOpsPlugin(), plugin_name="AzureOps")
+
+    return kernel
+
+
+async def get_chat_agent(kernel: Kernel) -> ChatCompletionAgent:
+    agent = ChatCompletionAgent(
+        kernel=kernel,
+        name="AzureOpsAgent",
+        instructions="You are an AI assistant that helps with Azure Operations. Use the available tools to analyze alerts and suggest fixes.",
+        function_choice_behavior=FunctionChoiceBehavior.Auto(),
+    )
+    return agent
+
+
+async def run_agent(query: str):
+    kernel = await get_agent()
+    agent = await get_chat_agent(kernel)
+
+    # Execute the agent
+    final_response = []
+    async for response in agent.invoke(query):
+        if response.content:
+            final_response.append(response.content)
+
+    return "".join(final_response)

src/backend/app.py

@@ -1,8 +1,42 @@
-from fastapi import FastAPI
+import logging
 
-app = FastAPI()
+from agent.azure_agent import run_agent
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
 
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
-@app.get("/")
-def read_root():
-    return {"Details": "Azure Agent Ops Copilot"}
+
+app = FastAPI(title="Azure Ops Copilot API")
+
+
+class ChatRequest(BaseModel):
+    message: str
+
+
+class ChatResponse(BaseModel):
+    response: str
+
+
+@app.post("/chat", response_model=ChatResponse)
+async def chat(request: ChatRequest):
+    try:
+        logger.info(f"Received chat request: {request.message}")
+        result = await run_agent(request.message)
+        return ChatResponse(response=str(result))
+    except Exception as e:
+        logger.error(f"Error processing request: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@app.get("/health")
+async def health():
+    return {"status": "healthy"}
+
+
+if __name__ == "__main__":
+    import uvicorn
+
+    uvicorn.run(app, host="0.0.0.0", port=8000)

src/backend/services/mcp_client.py

@@ -0,0 +1,10 @@
+from fastmcp import Client
+
+
+class MCPClient:
+    def __init__(self):
+        self.mcp_path = "src/mcp/main.py"
+        self.client = Client(self.mcp_path)
+
+    def execute(self, tool_name, **kwargs):
+        return self.client.call_tool(tool_name, **kwargs)

src/frontend/app.py

@@ -1,9 +1,36 @@
+import os
+
 import gradio as gr
+import requests
+
+API_URL = os.getenv("API_URL", "http://localhost:8000")
+
+
+def chat(message, history):
+    try:
+        response = requests.post(f"{API_URL}/chat", json={"message": message})
+        response.raise_for_status()
+        return response.json()["response"]
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+
+with gr.Blocks(title="Azure Ops Copilot") as demo:
+    gr.Markdown("# 🤖 Azure Ops Copilot")
+    gr.Markdown(
+        "Your AI assistant for Azure DevOps. Ask me to analyze alerts, check configs, or suggest fixes."
+    )
 
-app = gr.Blocks()
+    chatbot = gr.ChatInterface(
+        fn=chat,
+        examples=[
+            "Analyze alert alert-001",
+            "Check config for vm-01",
+            "Suggest a fix for high CPU on vm-01",
+        ],
+        title="Chat with Copilot",
+    )
 
-with app:
-    gr.Markdown("# Azure Agent Ops Copilot")
 
 if __name__ == "__main__":
-    app.launch()
+    demo.launch(server_name="0.0.0.0", server_port=7860)

src/mcp/Dockerfile

@@ -1 +0,0 @@
-FROM python:3.10-slim

src/mcp/main.py

@@ -0,0 +1,95 @@
+from fastmcp import FastMCP
+from utils import (
+    CONFIGS_FILE,
+    LOGS_FILE,
+    analyze_alert_logic,
+    generate_fix_logic,
+    get_resource_config_logic,
+)
+
+# Initialize FastMCP server
+mcp = FastMCP("Azure Ops Copilot")
+
+# --- Tools ---
+
+
+@mcp.tool()
+def analyze_alert(alert_id: str) -> str:
+    """
+    Analyze an Azure Monitor alert by ID.
+    Returns the alert details and potential root cause.
+    """
+    return analyze_alert_logic(alert_id)
+
+
+@mcp.tool()
+def get_resource_config(resource_id: str) -> str:
+    """
+    Get the configuration of an Azure resource.
+    """
+    return get_resource_config_logic(resource_id)
+
+
+@mcp.tool()
+def generate_fix(issue_type: str, resource_type: str) -> str:
+    """
+    Generate a fix for a specific issue type and resource type.
+    Returns a Bicep or CLI snippet.
+    """
+    return generate_fix_logic(issue_type, resource_type)
+
+
+@mcp.tool()
+def integration_placeholder(service_name: str, action: str) -> str:
+    """
+    Placeholder for future integrations (e.g., Jira, ServiceNow).
+    """
+    return (
+        f"Integration with {service_name} for action '{action}' is not yet implemented."
+    )
+
+
+# --- Resources ---
+
+
+@mcp.resource("azure://logs/recent")
+def get_recent_logs() -> str:
+    """Get the most recent Azure Monitor logs."""
+    if LOGS_FILE.exists():
+        return LOGS_FILE.read_text()
+    return "[]"
+
+
+@mcp.resource("azure://configs/all")
+def get_all_configs() -> str:
+    """Get all resource configurations."""
+    if CONFIGS_FILE.exists():
+        return CONFIGS_FILE.read_text()
+    return "[]"
+
+
+# --- Prompts ---
+
+
+@mcp.prompt()
+def analyze_issue(alert_id: str) -> str:
+    """Create a prompt to analyze an issue based on an alert ID."""
+    return f"""Please analyze the following alert and suggest remediation steps:
+    Alert ID: {alert_id}
+
+    1. Use the 'analyze_alert' tool to get details.
+    2. Use the 'get_resource_config' tool to check the resource configuration.
+    3. Use the 'generate_fix' tool if a fix is applicable.
+    """
+
+
+@mcp.prompt()
+def suggest_fix(resource_id: str, issue: str) -> str:
+    """Create a prompt to suggest a fix for a resource."""
+    return f"""The resource {resource_id} is experiencing {issue}.
+    Please generate a fix using the available templates.
+    """
+
+
+if __name__ == "__main__":
+    mcp.run()

src/mcp/server.py

@@ -1,13 +0,0 @@
-from fastmcp import FastMCP
-
-mcp = FastMCP("Azure AgentOps 🚀")
-
-
-@mcp.tool
-def add(a: int, b: int) -> int:
-    """Add two numbers"""
-    return a + b
-
-
-if __name__ == "__main__":
-    mcp.run()

src/mcp/tests/test_mcp.py

@@ -0,0 +1,46 @@
+import sys
+from pathlib import Path
+
+sys.path.append(str(Path(__file__).parent.parent))
+
+from src.mcp.logic import (
+    analyze_alert_logic,
+    generate_fix_logic,
+    get_resource_config_logic,
+)
+
+
+def test_analyze_alert():
+    print("Testing analyze_alert_logic...")
+    # Use a known ID from generation script logic (alert-001)
+    result = analyze_alert_logic("alert-001")
+    print(result)
+    assert "Alert Analysis" in result
+    assert "alert-001" in result
+
+
+def test_get_resource_config():
+    print("\nTesting get_resource_config_logic...")
+    # Use a known ID from generation script
+    resource_id = "/subscriptions/sub-1/resourceGroups/rg-1/providers/Microsoft.Compute/virtualMachines/vm-01"
+    result = get_resource_config_logic(resource_id)
+    print(result)
+    assert "Standard_D2s_v3" in result
+
+
+def test_generate_fix():
+    print("\nTesting generate_fix_logic...")
+    result = generate_fix_logic("High CPU", "Microsoft.Compute/virtualMachines")
+    print(result)
+    assert "vmSize" in result
+
+
+if __name__ == "__main__":
+    try:
+        test_analyze_alert()
+        test_get_resource_config()
+        test_generate_fix()
+        print("\nAll tests passed!")
+    except Exception as e:
+        print(f"\nTest failed: {e}")
+        sys.exit(1)