Starter AI Agents

Single-agent applications for learning agent fundamentals

Overview

Starter agents are single-agent applications that demonstrate core agent capabilities like tool use, reasoning, and task completion. These agents are perfect for learning agent development fundamentals before moving to more complex multi-agent systems.

Research & Analysis Agents

OpenAI Research Agent

A multi-agent research application that conducts comprehensive research on any topic using coordinated AI agents.

Features

• Triage Agent: Plans research approach and coordinates workflow
• Research Agent: Searches web and gathers relevant information
• Editor Agent: Compiles facts into comprehensive reports
• Automatic fact collection with source attribution
• Structured report generation with citations
• Interactive Streamlit UI with tracing

# Setup and run
git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
cd starter_ai_agents/openai_research_agent
pip install -r requirements.txt
export OPENAI_API_KEY='your-api-key-here'
streamlit run openai_researcher_agent.py

Research Process:

1. Enter research topic or select example
2. Triage agent plans the approach
3. Research agent gathers information from web
4. Editor agent compiles comprehensive report
5. View process in real-time and download report

AI Startup Trend Analysis Agent

Generates actionable insights by identifying nascent trends, market gaps, and growth opportunities in startup sectors.

Key Capabilities

• News collection using DuckDuckGo
• Article summarization with Newspaper4k
• Pattern identification in startup funding
• Technology adoption trend analysis
• Market opportunity detection
• Built with Claude 3.5 Sonnet

# Example usage pattern
# 1. Input specific startup sector (e.g., "AI healthcare")
# 2. Agent gathers recent news and funding data
# 3. Generates summaries of verified information
# 4. Identifies emerging patterns across stories

AI Reasoning Agent

Leverages advanced AI models for complex reasoning and decision-making tasks.

Features:

• Advanced reasoning with Ollama models
• Interactive playground interface
• Markdown-formatted outputs
• Customizable for different scenarios
• Local execution support

cd starter_ai_agents/ai_reasoning_agent
pip install -r requirements.txt
python local_ai_reasoning_agent.py

Data & Analysis Agents

AI Data Analysis Agent

An AI data analyst built with Agno and OpenAI’s GPT-4o that analyzes CSV and Excel files through natural language queries.

File Support

• CSV and Excel uploads
• Automatic schema inference
• Multiple file format support
• Data type detection

Analysis Features

• Natural language to SQL
• Complex aggregations
• Statistical summaries
• Data visualizations

# Quick start
cd starter_ai_agents/ai_data_analysis_agent
pip install -r requirements.txt
streamlit run ai_data_analyst.py

Powered by DuckDB:

# Example natural language queries:
# "What are the top 5 products by revenue?"
# "Show me the average sales by region"
# "Create a visualization of monthly trends"

No SQL knowledge required - the agent converts natural language questions into efficient SQL queries using DuckDB.

AI Data Visualization Agent

Your personal data visualization expert that generates charts and insights from natural language questions.

Multi-Model Support:

• Meta-Llama 3.1 405B (complex analysis)
• DeepSeek V3 (detailed insights)
• Qwen 2.5 7B (quick analysis)
• Meta-Llama 3.3 70B (advanced queries)

Capabilities:

• Automatic chart type selection
• Dynamic visualization generation
• Statistical analysis
• Custom plot formatting
• Interactive follow-up questions

Requires Together AI API key and E2B API key for sandbox execution.

Web & Automation Agents

Web Scraping AI Agent

AI-powered web scraping using ScrapeGraph AI - extract structured data with natural language prompts.

Local Library

Pros:

• Free to use (no API costs)
• Full control over execution
• Privacy-friendly (data stays local)
• Supports GPT-4o, GPT-5, local models

Cons:

• Requires local installation
• Limited by hardware
• Manual updates needed

cd starter_ai_agents/web_scraping_ai_agent
pip install -r requirements.txt
streamlit run ai_scrapper.py

Cloud SDK

Pros:

• No setup required (just API key)
• Scalable and fast
• Advanced features (SmartCrawler, SearchScraper)
• Always up-to-date

Cons:

• Pay-per-use (credit-based)
• Requires internet connection

cd scrapegraph_ai_sdk/
export SGAI_API_KEY='your-api-key'
python quickstart.py

Use Cases:

# E-commerce scraping
prompt = "Extract product names, prices, and availability"

# Content aggregation
prompt = "Extract article title, author, date, and main content"

# Competitive intelligence
prompt = "Extract pricing, features, and updates"

# Lead generation
prompt = "Find company names, emails, and phone numbers"

AI Meme Generator Agent

Browser automation agent that creates memes using AI and direct website interaction.

Multi-LLM Support:

• Claude 3.5 Sonnet (Anthropic)
• GPT-4o (OpenAI)
• Deepseek v3 (Deepseek)
• Automatic model switching

Browser Automation Features:

• Direct interaction with imgflip.com
• Automated template search
• Dynamic caption insertion
• Image link extraction
• Multi-step quality validation

cd starter_ai_agents/ai_meme_generator_agent_browseruse
pip install -r requirements.txt
python -m playwright install --with-deps
streamlit run ai_meme_generator_agent.py

Specialized Domain Agents

Medical Imaging Diagnosis Agent

AI-assisted analysis of medical images powered by Gemini 2.0 Flash built on Agno framework.

Analysis Components

Image Analysis:

• Image type identification (X-ray, MRI, CT, ultrasound)
• Anatomical region detection
• Key findings observation
• Abnormality detection
• Quality assessment

Diagnostic Output:

• Potential diagnoses ranking
• Differential diagnoses
• Severity assessment
• Patient-friendly explanations
• Visual reference points

cd starter_ai_agents/ai_medical_imaging_agent
pip install -r requirements.txt
streamlit run ai_medical_imaging.py

For educational purposes only. Not a replacement for professional medical diagnosis. Always consult qualified healthcare professionals.

Life Insurance Coverage Advisor Agent

Helps estimate term life insurance needs and surfaces available policy options.

Technology Stack:

• Agno agent framework
• OpenAI GPT-5 for reasoning
• E2B sandbox for calculations
• Firecrawl for web research

Features:

• Minimal intake form (age, income, dependents, debt, etc.)
• Python code execution in E2B sandbox
• Discounted cash-flow income replacement model
• Latest term-life product research
• Coverage breakdown with source links

cd starter_ai_agents/ai_life_insurance_advisor_agent
pip install -r requirements.txt
streamlit run life_insurance_advisor_agent.py

xAI Finance Agent

Financial analysis agent powered by xAI’s Grok model with real-time stock data.

Capabilities:

• Powered by Grok-4 Fast model
• Real-time stock data via YFinance
• Web search with DuckDuckGo
• Formatted tables for financial data
• Interactive playground interface

cd starter_ai_agents/xai_finance_agent
pip install -r requirements.txt
export XAI_API_KEY='your-api-key-here'
python xai_finance_agent.py

AgentOS Integration: Connect to AgentOS Control Plane for monitoring and management. See Connecting Your OS.

Multimodal Agents

Multimodal AI Agent

Combines video analysis and web search using Google’s Gemini 2.5 model.

Features:

• Video analysis (Gemini 2.5 Flash/Pro)
• Web research integration (DuckDuckGo)
• Multiple video formats (MP4, MOV, AVI)
• Real-time video processing
• Combined visual and textual analysis

cd starter_ai_agents/multimodal_ai_agent
pip install -r requirements.txt
export GOOGLE_API_KEY='your-api-key'
streamlit run multimodal_agent.py

Creative Content Agents

AI Music Generator Agent

Generates music using ModelsLab API and OpenAI’s GPT-4 for prompt optimization.

Features:

• Detailed prompt input (genre, instruments, mood)
• MP3 format output
• In-browser playback and download
• Simple Streamlit interface

cd starter_ai_agents/ai_music_generator_agent
pip install -r requirements.txt
streamlit run models_lab_music_generator_agent.py

Required API Keys:

• OpenAI API key
• ModelsLab API key

Blog to Podcast Agent

Converts blog posts into podcasts using GPT-4, Firecrawl, and ElevenLabs.

Workflow:

1. Blog scraping with Firecrawl API
2. Summary generation with GPT-4 (max 2000 chars)
3. Audio podcast creation with ElevenLabs
4. Playback and download capabilities

cd starter_ai_agents/ai_blog_to_podcast_agent
pip install -r requirements.txt
streamlit run blog_to_podcast_agent.py

Emotional Support Agents

Breakup Recovery Agent Team

Multi-agent system for emotional recovery support powered by Gemini 2.0 Flash.

Therapist Agent

Empathetic support and coping strategies with DuckDuckGo research tools.

Closure Agent

Writes unsent emotional messages for cathartic release.

Routine Planner

Daily recovery routines with balanced activities.

Brutal Honesty Agent

Direct, objective feedback with no sugar-coating.

Features:

• Chat screenshot analysis
• Parallel agent execution
• Secure API key management
• Team leader coordination

cd starter_ai_agents/ai_breakup_recovery_agent
pip install -r requirements.txt
streamlit run ai_breakup_recovery_agent.py

Getting Started Guide

1. Choose an Agent

   Select an agent based on your use case:

   • Research: OpenAI Research Agent
   • Data Analysis: Data Analysis Agent
   • Web Scraping: Web Scraping Agent
   • Specialized: Medical, Finance, or Insurance agents
2. Setup Environment

   # Clone repository
   git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
   cd starter_ai_agents/<agent-directory>
   
   # Install dependencies
   pip install -r requirements.txt
   
3. Configure API Keys

   Set required environment variables:

   export OPENAI_API_KEY='your-key'
   export GOOGLE_API_KEY='your-key'  # if needed
   export ANTHROPIC_API_KEY='your-key'  # if needed
   
4. Run the Agent

   # For Streamlit apps
   streamlit run <agent-file>.py
   
   # For CLI apps
   python <agent-file>.py
   

Common Patterns

Single Agent with Tools

from agno import Agent
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.yfinance import YFinanceTools

agent = Agent(
    name="Financial Analyst",
    role="Analyze stocks and market trends",
    tools=[DuckDuckGoTools(), YFinanceTools()],
    show_tool_calls=True,
    markdown=True
)

response = agent.run("What's the current price of AAPL?")

Agent with Memory

from agno.storage import SqlAgentStorage

agent = Agent(
    storage=SqlAgentStorage(
        table_name="agent_sessions",
        db_url="sqlite:///agent_data.db"
    ),
    add_history_to_messages=True
)

Streaming Responses

agent.print_response(
    "Analyze this data",
    stream=True
)

Start with research or data analysis agents to understand core patterns, then explore specialized agents for domain-specific implementations.

Next Steps

Advanced Agents

Explore more complex single-agent implementations

Multi-Agent Teams

Learn agent coordination and teamwork

Voice Agents

Add voice capabilities to your agents

MCP Agents

Integrate with external services via MCP