Beginner Overview

Welcome to Woodwork Engine — a compact, modular framework for building AI agents and workflows using a single declarative configuration file (commonly main.ww).

This short guide focuses on what beginners actually need to know to get started.

What you use as a beginner

• The woodwork CLI to start the program and interact with configured components.
• A configuration file (main.ww) that declares components (LLMs, inputs, outputs) and how they connect.
• A .env file for API keys and secrets referenced from the .ww config.

The basic flow for newcomers is: write a simple .ww config, add any required environment variables in .env, run woodwork --init to install runtime dependencies referenced by the config, then start woodwork and interact with your components.

Minimal mental model

1. Declare components in main.ww (for example: an LLM and a command-line input).
2. The parser reads main.ww and instantiates the configured components.
3. Run woodwork to start the CLI and send inputs to the configured components.

Note: internal orchestration details (internal modules/classes) may change over time — beginners do not need to understand those internals to use the tool.

Recommended next steps

1. Read docs/quickstart.md for a short install → run path.
2. Start with a single LLM and a command-line input in main.ww.
3. Explore examples/ for working configurations you can copy.

Where to contribute

• Add short, runnable tutorials to docs/tutorials/.
• Improve quickstart examples and .ww snippets to reduce friction for newcomers.

Thanks for improving the experience for new users — keeping docs short and focused helps people get productive faster.