Research Agent Example

This example demonstrates building an AI research agent that can investigate topics and provide cited summaries.

Enterprise-Ready Pattern

Uses 400+ modules including guardrails and enterprise compliance features. See Enterprise Documentation.

Overview

The research agent uses LLMs with guardrails to ensure factual, well-cited responses to research questions.

Key Features

• Web search and document retrieval
• Source citation requirements
• Content safety guardrails
• Performance monitoring

Code

```python from agenticaiframework.agents import Agent from agenticaiframework.tasks import Task from agenticaiframework.llms import LLMManager from agenticaiframework.guardrails import Guardrail from agenticaiframework.monitoring import Monitor

Example: AI Agent solving a research question

if name == "main": # Initialize components llm = LLMManager() llm.register_model("gpt-4", lambda prompt, kwargs: f"[Simulated GPT-4 Response to: {prompt}]") llm.set_active_model("gpt-4") guardrail = Guardrail(rules=["No harmful content", "Cite sources"]) monitor = Monitor()

# Create agent
research_agent = Agent(
    name="ResearchAgent",
    llm=llm,
    guardrail=guardrail,
    monitor=monitor
)

# Define task
research_task = Task(
    description="Research the impact of climate change on polar bear populations and summarize findings with citations.",
    expected_output="A concise, factual summary with at least 2 citations."
)

# Run task
result = research_agent.run_task(research_task)

# Output result
logger.info("=== Research Result ===")
logger.info(result)