Skip to main content

​
What are AI Agents?

AI agents are autonomous systems that use large language models to perceive, reason, and act to achieve specific goals. Unlike simple chatbots, agents can use tools, make decisions, and coordinate with other agents to complete complex tasks.

​
Agent Categories

Starter Agents

Single-agent applications perfect for learning agent fundamentals. Includes research, data analysis, web scraping, and specialized task agents.

Advanced Agents

Complex agent implementations with advanced capabilities including autonomous reasoning, medical imaging, and creative generation.

Multi-Agent Teams

Coordinated teams of specialized agents working together on complex workflows like legal analysis, recruitment, and travel planning.

Voice Agents

Voice-enabled agents with text-to-speech capabilities for customer support, audio tours, and conversational RAG systems.

MCP Agents

Agents built on the Model Context Protocol for seamless integration with GitHub, Notion, browsers, and multiple services.

Game Playing Agents

Autonomous agents that play strategic games like chess and tic-tac-toe using multi-agent architectures.

​
Key Capabilities

​
Tool Use

Agents can interact with external APIs, databases, and services to gather information and perform actions: 
# Example: Agent with web search and file system tools
from phidata import Agent
from phidata.tools.duckduckgo import DuckDuckGo
from phidata.tools.file import FileTools

agent = Agent(
    tools=[DuckDuckGo(), FileTools()],
    show_tool_calls=True
)

​
Memory and Context

Agents maintain conversation history and session state for coherent multi-turn interactions: 
# Persistent memory across sessions
agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    storage=SqlAgentStorage(
        table_name="agent_sessions",
        db_url="sqlite:///agent_memory.db"
    )
)

​
Multi-Agent Coordination

Multiple specialized agents collaborate on complex tasks: 
# Agent team with role specialization
researcher = Agent(
    role="Research Specialist",
    tools=[DuckDuckGo()]
)

analyst = Agent(
    role="Data Analyst",
    tools=[YFinanceTools()]
)

team_lead = Agent(
    role="Team Coordinator",
    team=[researcher, analyst]
)

​
Common Frameworks

The collection uses various agent frameworks:

Phidata

Build multi-modal agents with memory, knowledge, and tools. Used in finance, legal, and recruitment agents.

Agno

Modern agent framework with sandbox execution and tool integration. Powers coding and insurance agents.

OpenAI Agents SDK

Official SDK for building agents with OpenAI models. Used in voice and research agents.

Autogen

Microsoft’s framework for multi-agent conversations. Powers the chess game implementation.

​
Getting Started

1

Choose Your Agent Type

Start with Starter Agents to learn fundamentals, then progress to more complex implementations.
2

Set Up Your Environment

Most agents require API keys for LLM providers (OpenAI, Anthropic, Google) and service integrations.
export OPENAI_API_KEY='your-key-here'
pip install -r requirements.txt
3

Run Your First Agent

Clone the repository and run a simple agent:
git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
cd starter_ai_agents/openai_research_agent
streamlit run openai_researcher_agent.py
4

Customize and Extend

Modify agent instructions, add new tools, or combine multiple agents for your use case.

​
Use Cases by Domain

​
Business & Finance

​
Development & Research

​
Content & Creative

​
Healthcare & Professional

All agents in this collection are designed for educational purposes. Review API costs and rate limits before running agents at scale.

​
Next Steps

Explore Starter Agents

Begin with single-agent applications

Browse Agent Teams

See multi-agent coordination in action

Try Voice Agents

Build voice-enabled applications

Learn MCP Integration

Connect agents to external services

Build docs developers (and LLMs) love

Get started for freeTalk to us