Apache Training module: Building and running AI agents for open-source projects

This directory packages the maintainer-education stream as a reusable, LMS-neutral Apache Training module. Any project — ASF or not — can use it to teach the material in a structured course, not just circulate the pages as reading. It is shaped for upstream contribution to Apache Training so the module can live there once it is stable.

Who this is for

Instructors and facilitators running the education stream as a course: workshops, on-boarding sessions, reading groups, self-paced curricula, or any environment where learners need per-lesson structure (objectives, exercises, assessment) in addition to the reference pages.

Learners using the module self-paced do not need an LMS. Every lesson is a plain Markdown file: read it, work through the exercises, and answer the self-check questions before moving on.

Relationship to the source pages

Each lesson in this module is a wrapper around one or more progression pages. The source pages are the canonical reference; the lessons add LMS-friendly structure on top:

Learning objectives — what a learner will be able to do after the lesson, written as observable, assessable outcomes.
Exercises — hands-on activities a learner can complete without access to a live system (paper/whiteboard activities use the page's own examples).
Self-check — short questions (and answers) the learner uses to gate themselves before moving to the next lesson.

Nothing in this directory duplicates the reference material; it only frames it.

Module map

Lesson	Source page	Learning time
Lesson 1 — What agents are	What agents are	~30 min
Lesson 2 — Working with agents	Working with agents	~30 min
Lesson 3 — Choosing models	Choosing models	~35 min
Lesson 4 — Your first skill	Your first skill	~60 min
Lesson 5 — Writing safe skills	Writing safe skills	~45 min
Lesson 6 — Debugging a skill	Debugging a skill	~50 min
Lesson 7 — Writing portable skills	Writing portable skills	~35 min
Lesson 8 — Eval-driven development	Eval-driven development	~60 min
Lesson 9 — Agentic and autonomous work	Agentic and autonomous work	~45 min
Lesson 10 — English as a programming language	English as a programming language	~30 min
Lesson 11 — How to contribute	How to contribute	~30 min
Hands-on lab	Tutorial: build and evaluate a skill	~90 min

Delivery formats

Self-paced. Learners read the source page, then work through the lesson wrapper (objectives, exercises, self-check) on their own. No instructor or LMS needed.

Instructor-led. An instructor presents the key ideas from the source page, assigns the exercises to pairs or small groups, and uses the self-check questions for a brief group debrief before moving on. The facilitator guide (planned, not yet shipped) covers room setup, timing, and group discussion prompts.

LMS upload. Each lesson is a Markdown file that can be converted to SCORM, xAPI, or any other format a specific LMS supports. The module does not assume any particular LMS. Learning time estimates above are rough guides for LMS credit-hour tagging.

Prerequisites

No prior AI experience. Learners should be comfortable reading and writing plain text, and familiar with the idea of a software project that uses version control. Specific technical prerequisites are stated in each lesson.

Placeholders

Exercises use <PROJECT> wherever a real project name would appear. Substitute your own project name when working through the activities.

Licence

Apache License 2.0 (PRINCIPLE 17). Contributions carry a Generated-by: note in their commit message following ASF Generative Tooling Guidance.