# Install the SDK
pip install kubiya-workflow-sdk

# Set up environment
export TOGETHER_API_KEY="your-key"
export KUBIYA_API_KEY="your-key"

# Start the agent server
kubiya mcp agent --provider together --model "deepseek-ai/DeepSeek-V3" --port 8765

curl -X POST http://localhost:8765/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V3",
    "messages": [
      {"role": "user", "content": "Create a workflow that analyzes system performance"}
    ],
    "stream": true
  }'

import { useChat } from '@ai-sdk/react';

export function WorkflowChat() {
  const { messages, input, handleInputChange, handleSubmit } = useChat({
    api: 'http://localhost:8765/v1/chat/completions',
  });

  return (
    <div className="chat-container">
      {messages.map(m => (
        <div key={m.id} className={`message ${m.role}`}>
          {m.content}
        </div>
      ))}
      
      <form onSubmit={handleSubmit}>
        <input
          value={input}
          onChange={handleInputChange}
          placeholder="Describe the workflow you want to create..."
        />
        <button type="submit">Send</button>
      </form>
    </div>
  );
}

# Custom agent configuration
from kubiya_workflow_sdk.mcp import AgentServer

# Initialize with custom settings
server = AgentServer(
    provider="together",
    model="deepseek-ai/DeepSeek-V3",
    api_key="your-key"
)

# The server automatically:
# - Connects to MCP tools for workflow operations
# - Manages conversation context
# - Handles streaming responses
# - Executes workflows on Kubiya runners

User: "Create a workflow that backs up my database every night at 2 AM"

Agent: I'll create a scheduled database backup workflow for you...
[Generates and executes the workflow]

User: "Use Python to analyze this CSV file and create visualizations"

Agent: I'll create a data analysis workflow using Python...
[Creates workflow with python:3.11-slim image]

User: "Test my app in Python, Node, and Go simultaneously"

Agent: I'll set up parallel testing for all three languages...
[Creates workflow with parallel steps]

{"type": "step_running", "step": "build", "message": "Building Docker image..."}
{"type": "step_complete", "step": "build", "duration": "45s"}
{"type": "workflow_complete", "status": "success"}

# Set API keys
export KUBIYA_API_KEY="your-production-key"
export OPENAI_API_KEY="your-openai-key"

# Start with authentication
kubiya mcp agent --provider openai --port 8765

FROM python:3.11-slim

WORKDIR /app
RUN pip install kubiya-workflow-sdk

ENV KUBIYA_API_KEY=${KUBIYA_API_KEY}
ENV OPENAI_API_KEY=${OPENAI_API_KEY}

EXPOSE 8765
CMD ["kubiya", "mcp", "agent", "--provider", "openai", "--port", "8765"]

apiVersion: apps/v1
kind: Deployment
metadata:
  name: kubiya-agent
spec:
  replicas: 3
  selector:
    matchLabels:
      app: kubiya-agent
  template:
    metadata:
      labels:
        app: kubiya-agent
    spec:
      containers:
      - name: agent
        image: your-registry/kubiya-agent:latest
        ports:
        - containerPort: 8765
        env:
        - name: KUBIYA_API_KEY
          valueFrom:
            secretKeyRef:
              name: kubiya-secrets
              key: api-key

"Deploy my application with zero-downtime rolling update"
"Set up monitoring and alerting for my services"
"Create a disaster recovery workflow"

"Process CSV files and generate daily reports"
"Set up ETL pipeline for data warehouse"
"Analyze logs and detect anomalies"

"Create a full CI/CD pipeline with testing and deployment"
"Set up automated security scanning"
"Build multi-architecture Docker images"