How the academy uses the tools
Claude Code and Codex are the primary coding-agent workflows students practice. OpenClaw provides a hands-on lab surface for selected automation demos and proof artifacts.
Agentic AI course
Learn agentic AI by building one safe automation service offer.
This track frames agentic AI as a service skill: choose one repeated workflow, shape the assistant output, keep human approval visible, test the edge cases, and package a small offer a business owner can understand.
Service mapSafe demoTest notesOffer sentenceFinal project
Safety boundary
No autonomous customer-facing action is treated as launch-ready without sample-data testing, a human approval boundary, and clear client limitations.
Student artifacts
What students actually build.
A service map that names the business type, workflow pain, assistant output, and approval point.
A demo summary with sample data, output notes, and visible limitations.
A testing checklist for normal, missing-detail, risky, and not-a-fit cases.
A final project package that can be reviewed before any client promise is made.
Source-informed, not affiliated · non-affiliation note
Official product language shapes the course positioning.
These links are used for positioning and screenshot/source planning. The academy is independent educational training and is not officially affiliated with OpenAI, Anthropic, or OpenClaw.