Skip to main content
Navigate personas, integration models, and use cases for shipping PraisonAI recipes in production.
from praisonaiagents import Agent

agent = Agent(
    name="Recipe Guide",
    instructions="Point developers to the right recipe integration docs.",
)
agent.start("Where should I start integrating a customer-support recipe?")
The user picks a path below—models, personas, or use cases—to implement recipes in their app.

Quick Start

1

Simple Usage

List and run a recipe from the CLI.
praisonai recipe list
praisonai recipe run my-recipe --input '{"query": "Hello"}'
2

With Configuration

Serve recipes over HTTP so any language can call them.
praisonai serve recipe --port 8765

How It Works


Quick Navigation

Integration Models

6 ways to integrate recipes into your stack

Use Cases

12 real-world implementation patterns

Personas

Who uses recipes and how

Decision Guide

When to use which integration model

What Are Recipes?

Recipes are pre-packaged, reusable AI workflows that encapsulate:
  • Agent configurations
  • Tool assignments
  • Workflow logic
  • Input/output schemas
  • Security policies

Key Benefits

  • Reusability: Package once, deploy anywhere
  • Consistency: Same behavior across environments
  • Security: Built-in policy enforcement
  • Observability: Structured logging and tracing
  • Versioning: Track and rollback changes

Getting Started

# List available recipes
praisonai recipe list

# Run a recipe
praisonai recipe run my-recipe --input '{"query": "Hello"}'

# Start the recipe server
praisonai serve recipe --port 8765

Best Practices

Six models trade latency, language, and operational complexity. The Decision Guide maps your stack to the right one.
Recipes package agents, tools, and policies together. Keep them in source control so behaviour stays consistent across environments and rolls back cleanly.
praisonai serve recipe exposes recipes to any language via REST. Reach for the sidecar model when the caller is not Python.

Next Steps

  1. Review the Integration Models to choose your approach
  2. Explore Use Cases for implementation patterns
  3. Follow the step-by-step tutorials for your chosen model