Skip to main content

AgentMap Learning Guide

Welcome to the comprehensive AgentMap Learning Guide! This hands-on tutorial series takes you from basic workflow creation to advanced multi-agent orchestration through four progressive lessons.

Learning Path Overview​

Each lesson builds upon the previous one, introducing new concepts while reinforcing what you've learned. You'll work with real code examples and downloadable CSV files throughout the journey.

🎯 What You'll Learn​

  • Agent Fundamentals: Creating workflows with built-in agents
  • Prompt Engineering: Customizing AI behavior through effective prompting
  • Custom Agent Development: Building your own specialized agents
  • Multi-Agent Orchestration: Coordinating multiple agents intelligently

📈 Learning Progression​

Lesson Overview​

Lesson 1: Basic Agents​

Duration: 30-45 minutes | Difficulty: Beginner

Learn the fundamentals by creating a simple workflow that collects user input, processes it with AI, and saves the results to a CSV file.

Key Concepts:

  • InputAgent for user interaction
  • LLMAgent for AI processing
  • CSVWriterAgent for data persistence
  • Basic workflow structure

What You'll Build: A simple data collection and processing pipeline

Lesson 2: Custom Prompts​

Duration: 45-60 minutes | Difficulty: Beginner-Intermediate

Enhance your workflow by customizing AI behavior through prompt engineering and context parameters.

Key Concepts:

  • Prompt template design
  • Context parameters for LLMAgent
  • Temperature and model selection
  • Prompt optimization techniques

What You'll Build: An enhanced version of Lesson 1 with intelligent prompt customization

Lesson 3: Custom Agent Development​

Duration: 60-90 minutes | Difficulty: Intermediate

Create your own specialized agent by extending BaseAgent and implementing custom business logic.

Key Concepts:

  • BaseAgent inheritance patterns
  • Custom process() method implementation
  • CapableAgent protocols
  • Agent scaffold generation
  • Error handling and logging

What You'll Build: A custom validation agent with specialized processing logic

Lesson 4: Multi-Agent Orchestration​

Duration: 90-120 minutes | Difficulty: Intermediate-Advanced

Build a sophisticated workflow that intelligently routes requests between multiple specialized agents.

Key Concepts:

  • OrchestratorAgent usage
  • Keyword-based routing
  • Multiple custom agents
  • Advanced workflow patterns
  • Agent coordination strategies

What You'll Build: An intelligent routing system with multiple specialized processing agents

Prerequisites​

Technical Requirements​

  • Python 3.8 or higher
  • Basic Python programming knowledge
  • Text editor or IDE (VS Code recommended)
  • Terminal/command line familiarity

AgentMap Setup​

Ensure you have AgentMap installed and configured:

pip install agentmap
agentmap --version
  • Basic understanding of CSV files
  • Familiarity with JSON syntax
  • Basic command line usage
  • Understanding of Python dictionaries

Learning Resources​

Code Examples Repository​

All lesson files are available for download throughout the guide. Each lesson includes:

  • Complete CSV workflow files
  • Python agent code (where applicable)
  • Configuration examples
  • Test data samples

Getting Help​

Community Support​

  • GitHub Issues: Report bugs or ask technical questions
  • Documentation: Comprehensive reference materials
  • Examples: Real-world workflow patterns

Best Practices​

  • Start Simple: Complete each lesson in order
  • Experiment: Modify examples to understand concepts
  • Test Frequently: Run workflows after each change
  • Save Work: Keep your lesson files for reference

Success Tips​

1. Follow the Progression​

Each lesson builds on the previous one. Don't skip ahead - the concepts are designed to scaffold your learning.

2. Practice Variations​

Once you complete a lesson, try modifying the examples to deepen your understanding.

3. Use Real Data​

Replace sample data with your own use cases to make the learning more relevant.

4. Join the Community​

Share your progress and get help from other learners in the AgentMap community.

Ready to Start?​

Your learning journey begins with understanding the fundamentals. In Lesson 1, you'll create your first AgentMap workflow and see the power of chaining simple agents together.

Begin with Lesson 1: Basic Agents →

💡 Pro Tip: Keep a notebook handy to jot down insights and questions as you work through each lesson. The concepts you learn here will form the foundation for building sophisticated AI workflows.