Skip to content

Agent Execution Layer (AEL) Documentation

Welcome to the Agent Execution Layer documentation. AEL is a workflow execution engine with MCP (Model Context Protocol) integration that enables AI agents to execute complex, multi-step workflows.

What is AEL?

AEL (Agent Execution Layer) is a workflow execution engine that:

  • Executes YAML-defined workflows with tool and code steps
  • Integrates with MCP servers to access external tools
  • Provides a CLI for workflow management and execution
  • Exposes workflows as MCP tools for AI agent consumption
  • Supports REST API for programmatic access

Documentation Sections

Getting Started

Guide Description
Installation Install via pip, Docker, or from source
Quickstart 5-minute introduction to AEL
First Workflow Step-by-step workflow tutorial

Guides

Guide Description
Workflow Authoring Complete guide to writing workflows
Code Steps Using Python code in workflows
Tool Integration Connecting MCP tools
Troubleshooting Common issues and solutions

Reference

Reference Description
CLI Reference All CLI commands and options
Workflow Schema Complete YAML schema reference
Configuration Configuration file options
Error Codes Error codes and resolutions

Examples

Example Description
Web Scraping Extract data from websites
Data Processing Transform and process data
API Integration Integrate with external APIs

System Requirements

  • Python: 3.12 or higher
  • OS: macOS, Linux, Windows
  • Memory: 512MB minimum, 1GB recommended

Getting Help

License

AEL is released under the MIT License.