Modules 2–4 built a mental model of how LLMs work: next-token predictors, trained on internet text, shaped by RLHF. This lecture puts that model to work. Prompting is not a collection of magic phrases — it is an applied discipline grounded in how models process context. The principles covered here — clarity, positive framing, structure, chain-of-thought, and treating prompts as versioned code — apply directly to agent development, where prompt failures are not annoyances but bugs.
Claude Prompting Best Practices — Anthropic Docs — Anthropic's authoritative prompting reference covering clarity, XML structure, output formatting, and agentic use. Direct complement to this lecture.
Extended Thinking — Anthropic Docs — Claude's extended thinking API — the same CoT mechanism with a hidden reasoning block. Directly relevant for students building agents with Claude.
What is XML? — W3Schools — One-page intro: what XML is, what tags are, and how elements are structured. Good first read for students who haven't encountered XML before.
XML Syntax Rules — W3Schools — Covers proper nesting, closing tags, case sensitivity, and well-formedness — the mechanics that matter for understanding XML-style tags in prompts.
Basic Syntax — Markdown Guide — The most thorough beginner reference for Markdown, covering headers, lists, code blocks, and links with rendered examples.
Cheat Sheet — Markdown Guide — Compact one-page summary of all common Markdown syntax. Best as a quick lookup during the lab.
CommonMark Help — Interactive reference for CommonMark, the standardized Markdown spec used by GitHub and others.