Documentation Index
Fetch the complete documentation index at: https://docs.praison.ai/llms.txt
Use this file to discover all available pages before exploring further.
Auto Module
The Auto module provides automated agent generation from natural language prompts.
Installation
Quick Start
from praisonai import PraisonAI
# Generate agents from a prompt
praison = PraisonAI(auto="Create a team to research AI trends")
result = praison.run()
CLI Usage
# Auto mode with prompt
praisonai --auto "Create a research team for market analysis"
# Auto mode with topic
praisonai --auto "Build a content creation pipeline"
How It Works
- Prompt Analysis: The auto module analyzes your natural language prompt
- Agent Generation: Creates appropriate agents with roles, goals, and tasks
- YAML Creation: Generates a valid agents.yaml configuration
- Execution: Runs the generated agents
Configuration
Environment Variables
| Variable | Description |
|---|
OPENAI_API_KEY | OpenAI API key for agent generation |
PRAISONAI_MODEL | Model to use for generation (default: gpt-4o-mini) |
Generated Structure
Auto mode generates a YAML structure like:
framework: praisonaiagents
topic: Research AI trends
roles:
researcher:
role: Research Analyst
goal: Research and analyze AI trends
backstory: Expert researcher with analytical skills
tasks:
research_task:
description: Research the latest AI developments
expected_output: Comprehensive research report
writer:
role: Content Writer
goal: Create engaging content from research
backstory: Skilled writer with technical background
tasks:
write_task:
description: Write article based on research
expected_output: Well-structured article
Programmatic Usage
from praisonai import PraisonAI
# Basic auto generation
praison = PraisonAI(auto="Create a customer support team")
result = praison.run()
# With specific framework
praison = PraisonAI(
auto="Build a data analysis pipeline",
framework="praisonaiagents"
)
result = praison.run()
Customization
from praisonai import PraisonAI
def search_tool(query: str) -> str:
return f"Search results for: {query}"
praison = PraisonAI(
auto="Create a research team with web search",
tools=[search_tool]
)
result = praison.run()
See Also
