Learn Agentic Coding with Codex and AI Coding Agents

By Beau Carnes | November 10, 2026

AI coding agents such as Codex can read your codebase, edit multiple files, and run commands and tests while collaborating with you on larger tasks. In Codex, an AGENTS.md file stores persistent project instructions, while config.toml holds the agent's configuration. As agentic coding matures in 2026, mastering these building blocks is essential for getting dependable, reproducible results.

We just published a course on the freeCodeCamp.org YouTube channel that teaches you how to get reliable results from a coding agent by giving it the right context, planning the work, controlling its permissions, and verifying every change. The course is taught by Tom Chant, Head of Education at Scrimba.

You'll work through realistic tasks, including challenges on skills, characterization tests, and config.toml. The course is aimed at experienced developers adding AI to their workflow and at non-coders who want to build software with AI assistance. It is also available as an interactive course on Scrimba.

What You'll Learn

• Prompting & Project Setup — Write effective prompts and comment prompts for a coding agent, then initialize a project and work with multiple AGENTS.md files.
• Plan Mode & Skills — Break larger development tasks down with Plan Mode and extend a coding agent with reusable skills.
• Testing Agent Changes — Verify an agent's changes with characterization tests and test-driven development (TDD), and learn which testing pitfalls to watch for.
• Git & Secrets — Use Git safely alongside a coding agent and keep API keys, credentials, and other secrets secure.
• Context, Permissions & config.toml — Use the VS Code Codex extension, manage context, control what an agent can access and change, and configure it with config.toml.
• MCP & shadcn — Connect agents to external tools with the Model Context Protocol (MCP), including shadcn integrations.

Watch the Course

Watch the full course on the freeCodeCamp.org YouTube channel (approximately 2 hours).

via FreeCodeCamp