Skip to main content

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.

Research agents search the web, gather information, and provide summarized insights on any topic.

Quick Start

from praisonaiagents import Agent

# Research agent with web search
researcher = Agent(
    name="Researcher",
    instructions="Research topics and provide clear summaries",
    web=True  # Enable web search
)

researcher.start("What are the latest trends in AI?")
Enable web=True to give your agent web search capabilities.

How It Works


Complete Example

from praisonaiagents import Agent

# Create a research agent
researcher = Agent(
    name="ResearchAgent",
    instructions="""You are a research specialist.

When researching:
- Search for accurate, current information
- Organize findings clearly
- Identify key facts and insights
- Present in a structured format""",
    web=True
)

# Research a topic
researcher.start("Research the impact of renewable energy on carbon emissions")

Structured Research Output

from praisonaiagents import Agent

researcher = Agent(
    name="StructuredResearcher",
    instructions="""You research topics and provide structured reports.

Format your response as:
## Summary
Brief overview

## Key Facts
- Fact 1
- Fact 2
- Fact 3

## Analysis
Deeper insights

## Sources
List of sources""",
    web=True
)

researcher.start("Research quantum computing advancements")

Specialized Research Agents

Comparison Agent

Compare options side-by-side

Trend Analyst

Identify patterns and trends

Fact Checker

Verify claims and statements

Literature Review

Summarize academic research

Comparison Agent

comparison_agent = Agent(
    name="ComparisonAgent",
    instructions="""Compare options and present findings.
    
- Identify key criteria
- Research each option
- Create side-by-side comparison
- Highlight pros and cons""",
    web=True
)

comparison_agent.start("Compare solar vs wind energy for homes")

Research Team

Multiple agents working together for deep research:
from praisonaiagents import Agent, AgentTeam

# Researcher finds information
researcher = Agent(
    name="Researcher",
    instructions="Find comprehensive information on topics",
    web=True
)

# Analyst interprets findings
analyst = Agent(
    name="Analyst",
    instructions="Analyze research and identify key insights"
)

# Writer creates the report
writer = Agent(
    name="Writer",
    instructions="Write clear, structured research reports"
)

# Create research team
team = AgentTeam(
    agents=[researcher, analyst, writer],
    process="sequential"
)

team.start("Research the future of electric vehicles")

Best Practices

Be specific about what you want to research
Ask for organized output with sections
Use multiple sources for important claims
Combine research + analysis + writing agents

Next: Content Creation Agents

Learn how to build agents that create content.