Agent MAX Examples

Agent MAX is currently in private beta. These examples show what’s possible. Join the waitlist to get access.

These aren’t toy demos. These are production-ready patterns for real tasks.

Lead Research

Company research with contact enrichment

Churn Analysis

Identify at-risk customers, draft outreach

Contract Review

Extract risks from legal documents

Competitor Monitoring

Track changes across competitor websites

Meeting Prep

Automatic briefing documents

Data Migration

Transform and validate large datasets

Lead Research

Goal: Given a company name, produce a complete research report with decision makers and contact info. Time: ~30 seconds
Traditional approach: ~5 minutes of manual searching

from incredible import AgentMax

agent = AgentMax(api_key="YOUR_API_KEY")

result = agent.run_with_results(
    goal="""
    Research the provided company thoroughly:
    1. Company overview (size, industry, funding)
    2. Recent news and announcements
    3. Key decision makers (C-suite, VPs)
    4. Tech stack if available
    5. Potential pain points based on their industry
    
    Output a structured research brief I can use for outreach.
    """,
    data={
        "company": {
            "name": "Stripe",
            "website": "stripe.com",
            "industry": "fintech"
        }
    },
    tools=[
        web_search,
        linkedin_search,
        crunchbase_api,
        builtwith_api  # tech stack detection
    ],
    result_structure={
        "type": "object",
        "properties": {
            "company_overview": {"type": "object"},
            "recent_news": {"type": "array"},
            "decision_makers": {"type": "array"},
            "tech_stack": {"type": "array"},
            "pain_points": {"type": "array"}
        }
    }
)

print(result.output)

Sample Output (result.output):

{
  "company_overview": {
    "name": "Stripe",
    "founded": 2010,
    "headquarters": "San Francisco, CA",
    "employees": 8000,
    "valuation": "50B",
    "industry": "Fintech / Payment Processing"
  },
  "recent_news": [
    "Launched Stripe Billing v3 (March 2024)",
    "Expanded to 5 new LATAM countries",
    "Partnership with Shopify deepened"
  ],
  "decision_makers": [
    { "name": "Patrick Collison", "title": "CEO", "linkedin": "linkedin.com/in/patrickcollison" },
    { "name": "Claire Hughes Johnson", "title": "COO", "linkedin": "linkedin.com/in/chughes" },
    { "name": "David Singleton", "title": "CTO", "linkedin": "linkedin.com/in/dsingleton" }
  ],
  "tech_stack": ["React", "Ruby on Rails", "AWS", "Kubernetes", "Kafka"],
  "pain_points": [
    "Scaling support for enterprise clients",
    "Regulatory compliance in new markets",
    "Competition from Adyen in EU market"
  ]
}

Churn Analysis

Goal: Analyze customer data, identify churn risk, draft personalized outreach. Input: CSV with customer data
Output: Risk scores + ready-to-send email drafts

from incredible import AgentMax

agent = AgentMax(api_key="YOUR_API_KEY")

# Open files properly
with open("customers.csv", "r") as customers, \
     open("support_tickets.csv", "r") as tickets, \
     open("usage_logs.csv", "r") as logs:
    
    result = agent.run_with_results(
        goal="""
        Analyze the customer data to identify churn risk:
        
        1. Calculate engagement metrics (login frequency, feature usage)
        2. Identify customers with declining engagement
        3. Cross-reference with support ticket sentiment
        4. Score each customer 0-100 for churn risk
        5. For top 20 at-risk customers, draft personalized outreach emails
        
        Focus on customers with >$10k ARR first.
        """,
        files=[customers, tickets, logs],
        result_structure={
            "type": "object",
            "properties": {
                "summary": {
                    "type": "object",
                    "properties": {
                        "total_analyzed": {"type": "integer"},
                        "high_risk_count": {"type": "integer"},
                        "total_arr_at_risk": {"type": "number"}
                    }
                },
                "at_risk_customers": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "customer_id": {"type": "string"},
                            "company_name": {"type": "string"},
                            "arr": {"type": "number"},
                            "risk_score": {"type": "integer"},
                            "risk_factors": {"type": "array", "items": {"type": "string"}},
                            "draft_email": {"type": "string"}
                        }
                    }
                }
            }
        }
    )

# Output is guaranteed to match the schema
for customer in result.output["at_risk_customers"][:5]:
    print(f"\n{'='*50}")
    print(f"🚨 {customer['company_name']} - Risk: {customer['risk_score']}/100")
    print(f"   ARR: ${customer['arr']:,.0f}")
    print(f"   Factors: {', '.join(customer['risk_factors'])}")
    print(f"\n   📧 Draft email:\n   {customer['draft_email'][:200]}...")

Contract Review

Goal: Extract risk factors from legal documents with specific clause references. Input: PDF contract
Output: Structured risk assessment

from incredible import AgentMax

agent = AgentMax(api_key="YOUR_API_KEY")

with open("vendor_contract.pdf", "rb") as contract:
    result = agent.run_with_results(
        goal="""
        Review this contract and identify potential risks:
        
        1. Liability clauses (caps, indemnification)
        2. Termination conditions
        3. Auto-renewal terms
        4. Data handling requirements
        5. Unusual or non-standard clauses
        
        For each risk, cite the specific section and page number.
        Rate overall contract risk: Low / Medium / High
        """,
        files=[contract],
        result_structure={
            "type": "object",
            "properties": {
                "overall_risk": {"type": "string", "enum": ["Low", "Medium", "High"]},
                "summary": {"type": "string"},
                "risks": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "category": {"type": "string"},
                            "description": {"type": "string"},
                            "severity": {"type": "string"},
                            "section": {"type": "string"},
                            "page": {"type": "integer"},
                            "recommendation": {"type": "string"}
                        }
                    }
                },
                "key_dates": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "event": {"type": "string"},
                            "date": {"type": "string"}
                        }
                    }
                }
            }
        }
    )

print(f"Overall Risk: {result.output['overall_risk']}")
print(f"\nSummary: {result.output['summary']}")
print(f"\nKey Risks Found: {len(result.output['risks'])}")

for risk in result.output['risks']:
    print(f"\n⚠️  {risk['category']} ({risk['severity']})")
    print(f"   {risk['description']}")
    print(f"   📍 Section {risk['section']}, Page {risk['page']}")
    print(f"   💡 {risk['recommendation']}")

Competitor Monitoring

Goal: Track competitor changes and generate daily digest. Setup: Runs on schedule (daily/weekly)
Output: Email digest with changes

from incredible import AgentMax

agent = AgentMax(api_key="YOUR_API_KEY")

# Define competitors to track
competitors = [
    {"name": "Competitor A", "website": "competitora.com", "pricing_page": "competitora.com/pricing"},
    {"name": "Competitor B", "website": "competitorb.com", "pricing_page": "competitorb.com/plans"},
]

result = agent.run_with_results(
    goal="""
    Monitor competitors for changes since yesterday:
    
    1. Check pricing pages for any changes
    2. Look for new blog posts or announcements
    3. Check for new features mentioned on marketing pages
    4. Search news for mentions or press releases
    5. Check their job postings for strategic hires
    
    Compare against yesterday's snapshot and highlight what's new.
    """,
    data={
        "competitors": competitors,
        "last_check": "2024-03-14"
    },
    tools=[
        web_search,
        page_snapshot,
        diff_checker,
        news_search
    ],
    result_structure={
        "type": "object",
        "properties": {
            "digest_date": {"type": "string"},
            "competitors": {
                "type": "array",
                "items": {
                    "type": "object",
                    "properties": {
                        "name": {"type": "string"},
                        "changes_detected": {"type": "boolean"},
                        "pricing_changes": {"type": "array"},
                        "new_content": {"type": "array"},
                        "job_postings": {"type": "array"}
                    }
                }
            }
        }
    }
)

# Format and send the digest
digest = format_digest(result.output)
send_email(
    to="[email protected]",
    subject="🔍 Daily Competitor Digest - March 15, 2024",
    body=digest
)

Sample Output (result.output):

{
  "digest_date": "2024-03-15",
  "competitors": [
    {
      "name": "Competitor A",
      "changes_detected": true,
      "pricing_changes": [
        "Enterprise tier increased from $499 → $599/month",
        "New 'Startup' tier added at $99/month"
      ],
      "new_content": [
        "Blog: 'How We Scaled to 1M Users' - mentions AWS partnership"
      ],
      "job_postings": [
        "Head of AI/ML"
      ]
    },
    {
      "name": "Competitor B",
      "changes_detected": false,
      "pricing_changes": [],
      "new_content": [],
      "job_postings": []
    }
  ]
}

Formatted Digest (after format_digest()):

# 🔍 Competitor Digest - March 15, 2024

## Competitor A
### 🆕 Changes Detected

**Pricing Change** ⚠️
- Enterprise tier increased from $499 → $599/month
- New "Startup" tier added at $99/month

**New Blog Post**
- "How We Scaled to 1M Users" - mentions AWS partnership

**Job Posting**
- Hiring: "Head of AI/ML" - Suggests ML product investment

---

## Competitor B
### ✅ No significant changes

Meeting Prep

Goal: Generate a briefing document before any meeting. Trigger: Calendar event with external attendees
Output: Research brief in your inbox

from incredible import AgentMax
from datetime import timedelta

agent = AgentMax(api_key="YOUR_API_KEY")

# Triggered by calendar integration
meeting = {
    "title": "Partnership Discussion",
    "attendees": ["[email protected]", "[email protected]"],
    "company": "TechStartup Inc",
    "time": "2024-03-15 14:00"
}

result = agent.run_with_results(
    goal="""
    Prepare a meeting brief for my upcoming call:
    
    1. Research the company (TechStartup Inc)
       - What they do, size, funding stage
       - Recent news or announcements
    
    2. Research the attendees
       - John and Sarah's roles and backgrounds
       - Any mutual connections
       - Their recent LinkedIn activity/posts
    
    3. Prepare context
       - Any previous interactions in our CRM
       - Open deals or past conversations
       - Support tickets or issues
    
    4. Suggest talking points and questions to ask
    """,
    data={"meeting": meeting},
    tools=[
        web_search,
        linkedin_search,
        crm_search,
        email_search
    ],
    result_structure={
        "type": "object",
        "properties": {
            "company_info": {"type": "object"},
            "attendees": {"type": "array"},
            "previous_interactions": {"type": "array"},
            "talking_points": {"type": "array"},
            "questions_to_ask": {"type": "array"}
        }
    }
)

# Format and send to my email 30 min before meeting
brief = format_meeting_brief(result.output)
send_email(
    to="[email protected]",
    subject=f"📋 Meeting Brief: {meeting['title']} with {meeting['company']}",
    body=brief,
    send_at=meeting['time'] - timedelta(minutes=30)
)

Data Migration

Goal: Transform data from one format to another with validation. Input: Export from legacy system
Output: Clean, validated data for new system

from incredible import AgentMax
import json

agent = AgentMax(api_key="YOUR_API_KEY")

with open("legacy_export.csv", "r") as legacy_file:
    result = agent.run_with_results(
        goal="""
        Migrate customer data from the legacy format to our new schema:
        
        1. Parse the legacy CSV export
        2. Map fields to new schema:
           - "cust_name" → "company_name"
           - "contact_1" → "primary_contact.name"
           - "email_1" → "primary_contact.email"
           - etc.
        3. Validate all email addresses
        4. Standardize phone numbers to E.164 format
        5. Deduplicate based on email domain
        6. Flag records that need manual review
        
        Output the migrated data and a summary of issues.
        """,
        files=[legacy_file],
        result_structure={
            "type": "object",
            "properties": {
                "migrated_records": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "company_name": {"type": "string"},
                            "primary_contact": {
                                "type": "object",
                                "properties": {
                                    "name": {"type": "string"},
                                    "email": {"type": "string"},
                                    "phone": {"type": "string"}
                                }
                            },
                            "status": {"type": "string"}
                        }
                    }
                },
                "summary": {
                    "type": "object",
                    "properties": {
                        "total_input": {"type": "integer"},
                        "successfully_migrated": {"type": "integer"},
                        "duplicates_removed": {"type": "integer"},
                        "needs_review": {"type": "integer"}
                    }
                },
                "issues": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "row": {"type": "integer"},
                            "field": {"type": "string"},
                            "issue": {"type": "string"},
                            "original_value": {"type": "string"}
                        }
                    }
                }
            }
        }
    )

print(f"Migration Summary:")
print(f"  Input records: {result.output['summary']['total_input']}")
print(f"  Migrated: {result.output['summary']['successfully_migrated']}")
print(f"  Duplicates removed: {result.output['summary']['duplicates_removed']}")
print(f"  Needs review: {result.output['summary']['needs_review']}")

# Export for review
with open('migration_issues.json', 'w') as f:
    json.dump(result.output['issues'], f, indent=2)

Run Your Own

Ready to build something? Start with the SDK:

pip install incredible-python

SDK Reference

Complete API documentation

Agent MAX

​Agent MAX Examples

Lead Research

Churn Analysis

Contract Review

Competitor Monitoring

Meeting Prep

Data Migration

​Lead Research

​Churn Analysis

​Contract Review

​Competitor Monitoring

​Meeting Prep

​Data Migration

​Run Your Own

SDK Reference

Agent MAX Examples

Lead Research

Churn Analysis

Contract Review

Competitor Monitoring

Meeting Prep

Data Migration

Run Your Own