Code Examples

import { FantomuClient } from '@fantomu/sdk';

const client = new FantomuClient({
  apiKey: process.env.FANTOMU_API_KEY
});

// Create email assistant agent
const emailAgent = await client.agents.create({
  name: 'Email Assistant',
  description: 'AI agent for email management and drafting',
  capabilities: [
    'email_drafting',
    'email_scheduling', 
    'email_analysis',
    'contact_management'
  ]
});

// Draft a professional email
const emailDraft = await emailAgent.execute({
  task: 'Draft a professional follow-up email',
  context: {
    recipient: 'client@example.com',
    subject: 'Project Update - Q1 Review',
    tone: 'professional',
    previousEmails: 2,
    meetingDate: '2024-01-15'
  }
});

console.log('Email Draft:', emailDraft.result);

// Schedule email for later
await emailAgent.execute({
  task: 'Schedule this email to be sent tomorrow at 9 AM',
  context: {
    emailContent: emailDraft.result,
    sendTime: '2024-01-16T09:00:00Z'
  }
});

from fantomu import FantomuClient
import pandas as pd
import json

client = FantomuClient(api_key="your-api-key")

# Create data analysis agent
analysis_agent = client.agents.create(
    name="Data Analysis Assistant",
    description="AI agent for data analysis and insights generation",
    capabilities=[
        "data_analysis",
        "statistical_analysis", 
        "visualization_recommendations",
        "insight_generation"
    ]
)

# Load your dataset
df = pd.read_csv('sales_data.csv')

# Analyze the data
analysis_result = analysis_agent.execute(
    task="Analyze this sales data and provide key insights",
    context={
        "data_summary": {
            "rows": len(df),
            "columns": list(df.columns),
            "date_range": f"{df['date'].min()} to {df['date'].max()}"
        },
        "sample_data": df.head().to_dict(),
        "analysis_type": "sales_performance"
    }
)

print("Analysis Results:")
print(analysis_result.result)

# Generate visualization recommendations
viz_recommendations = analysis_agent.execute(
    task="Recommend the best visualizations for this data",
    context={
        "data_columns": list(df.columns),
        "analysis_insights": analysis_result.result
    }
)

print("\nVisualization Recommendations:")
print(viz_recommendations.result)

import { FantomuClient } from '@fantomu/sdk';

const client = new FantomuClient({
  apiKey: process.env.FANTOMU_API_KEY
});

// Create customer support agent
const supportAgent = await client.agents.create({
  name: 'Customer Support Bot',
  description: 'AI agent for handling customer inquiries and support tickets',
  capabilities: [
    'ticket_classification',
    'response_generation',
    'escalation_detection',
    'knowledge_base_search'
  ]
});

// Handle incoming support ticket
async function handleSupportTicket(ticket) {
  // Classify the ticket
  const classification = await supportAgent.execute({
    task: 'Classify this support ticket by priority and category',
    context: {
      ticketId: ticket.id,
      subject: ticket.subject,
      description: ticket.description,
      customerTier: ticket.customerTier
    }
  });

  console.log('Ticket Classification:', classification.result);

  // Generate response based on classification
  const response = await supportAgent.execute({
    task: 'Generate an appropriate response for this support ticket',
    context: {
      ticketClassification: classification.result,
      ticketDetails: ticket,
      knowledgeBase: await getKnowledgeBase(),
      escalationThreshold: 'high_priority'
    }
  });

  // Check if escalation is needed
  const escalationCheck = await supportAgent.execute({
    task: 'Determine if this ticket needs human escalation',
    context: {
      ticketClassification: classification.result,
      response: response.result,
      customerSatisfaction: ticket.customerSatisfaction
    }
  });

  return {
    ticketId: ticket.id,
    classification: classification.result,
    response: response.result,
    needsEscalation: escalationCheck.result.needsEscalation
  };
}

// Example usage
const ticket = {
  id: 'TICKET-12345',
  subject: 'Unable to access my account',
  description: 'I keep getting an error when trying to log in...',
  customerTier: 'premium'
};

const result = await handleSupportTicket(ticket);
console.log('Support Ticket Result:', result);

from fantomu import FantomuClient
import asyncio

client = FantomuClient(api_key="your-api-key")

# Create specialized agents
research_agent = client.agents.create(
    name="Research Agent",
    description="Agent specialized in research and fact-checking",
    capabilities=["web_research", "fact_checking", "source_verification"]
)

writer_agent = client.agents.create(
    name="Content Writer",
    description="Agent specialized in content writing and editing",
    capabilities=["content_writing", "editing", "seo_optimization"]
)

editor_agent = client.agents.create(
    name="Editor Agent", 
    description="Agent specialized in proofreading and quality control",
    capabilities=["proofreading", "grammar_check", "style_consistency"]
)

async def create_content_pipeline(topic, requirements):
    """Multi-agent content creation pipeline"""
    
    # Step 1: Research
    research_result = await research_agent.execute(
        task="Research comprehensive information about this topic",
        context={
            "topic": topic,
            "requirements": requirements,
            "sources_needed": 5,
            "fact_check_required": True
        }
    )
    
    print("Research completed:", research_result.result)
    
    # Step 2: Content Writing
    content_result = await writer_agent.execute(
        task="Write high-quality content based on the research",
        context={
            "topic": topic,
            "research_data": research_result.result,
            "requirements": requirements,
            "target_audience": requirements.get("audience", "general"),
            "content_type": requirements.get("type", "article")
        }
    )
    
    print("Content written:", content_result.result)
    
    # Step 3: Editing and Quality Control
    edit_result = await editor_agent.execute(
        task="Edit and improve the content for quality and consistency",
        context={
            "original_content": content_result.result,
            "research_sources": research_result.result,
            "style_guide": requirements.get("style_guide", "professional"),
            "word_count_target": requirements.get("word_count", 1000)
        }
    )
    
    print("Content edited:", edit_result.result)
    
    return {
        "research": research_result.result,
        "draft": content_result.result,
        "final": edit_result.result
    }

# Example usage
topic = "The Future of AI in Healthcare"
requirements = {
    "type": "blog_post",
    "audience": "healthcare_professionals",
    "word_count": 1500,
    "style_guide": "professional"
}

result = await create_content_pipeline(topic, requirements)
print("Content Pipeline Complete!")