Industry Templates

Pre-built agent workforces for Manufacturing, Energy, Healthcare, Agriculture, and Transportation — ready to run in minutes.

Five production-ready agent pipelines built on the SRAO Framework, each targeting a specific industry with typed I/O schemas, SLA targets, and graceful fallback strategies.

Five production-ready agent pipelines built on the SRAO Framework, each targeting a specific industry with typed I/O schemas, SLA targets, and graceful fallback strategies.

from praisonaiagents import Agent
from examples.cookbooks.Industry_Templates.manufacturing_template import manufacturing_workflow

agent = Agent(name="ops-lead", instructions="Route industry template workflows.")
result = manufacturing_workflow(
    "Customer ABC needs 100 units of Product-XYZ by March 15th, high priority"
)
print(result)

The user picks an industry template; agents parse the request and return a structured operations report.

Quick Start

1. Pick an industry and run

   Each template ships a single workflow function you can call immediately:

   from praisonaiagents import Agent, tool
   
   # Manufacturing: order → inventory → schedule → quality
   from examples.cookbooks.Industry_Templates.manufacturing_template import manufacturing_workflow
   
   result = manufacturing_workflow("Customer ABC needs 100 units of Product-XYZ by March 15th, high priority")
   print(result)

2. Customise with IndustryAgentPattern

   Every template exposes IndustryAgentPattern — a base class you can mix into any domain:

   from examples.cookbooks.Industry_Templates.manufacturing_template import IndustryAgentPattern
   
   # Build a domain-specific parser in one line
   parser = IndustryAgentPattern.create_data_parser(
       name="CustomParser",
       domain="pharmaceutical batch records",
       sla_seconds=20
   )
   
   result = parser.start("Parse batch record BR-2024-001")
   print(result)

Which Template Should I Use?

Templates at a Glance

How It Works

All five templates share the same SRAO architecture — only the domain logic differs.

Every step includes a fallback strategy — if an agent fails, the workflow degrades gracefully rather than crashing.

Cross-Industry Reuse with IndustryAgentPattern

IndustryAgentPattern provides four factory methods that power ~70 % of every template:

from examples.cookbooks.Industry_Templates.manufacturing_template import IndustryAgentPattern

# Generic parser — works for any domain
parser    = IndustryAgentPattern.create_data_parser("MyParser", "logistics data", sla_seconds=15)

# Generic availability checker
checker   = IndustryAgentPattern.create_resource_checker("BedChecker", "hospital bed", sla_seconds=3)

# Generic optimiser
optimizer = IndustryAgentPattern.create_optimizer("IrrigationOpt", "water usage", sla_minutes=5)

# Generic inspector
inspector = IndustryAgentPattern.create_inspector("TunnelInspect", "structural integrity", sla_minutes=2)

Pydantic I/O Schemas

Every template uses Pydantic models to enforce typed data between agents:

Common Patterns

Add a custom tool to any agent

from praisonaiagents import Agent, tool

@tool
def my_erp_lookup(order_id: str) -> dict:
    """Look up order in ERP system"""
    return {"erp_status": "confirmed", "order_id": order_id}

from examples.cookbooks.Industry_Templates.manufacturing_template import order_parser

order_parser.tools.append(my_erp_lookup)
result = order_parser.start("Parse order ORD-999")

Override the fallback strategy

def custom_fallback(error: Exception) -> dict:
    return {"status": "queued_for_review", "error": str(error)}

try:
    result = manufacturing_workflow("urgent order text")
except Exception as e:
    result = custom_fallback(e)

Best Practices

Related