How to Prompt.

You will learn how current Claude models follow instructions, and the five fixes that turn vague prompts into precise ones. End state is a rewritten version of your own worst prompt, tested side by side against the original. After this module you can scope outputs explicitly, cap length and format, use positive instructions and action verbs, and know when to ask for extra reasoning.

• Anthropic, "Prompt engineering overview", the techniques that hold across every current model: https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/overview
• Anthropic, "Prompting Claude Opus 5", the source for the response-length reversal and for leaving self-verification to the model: https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5
• Claude Help Center, "Change the model effort and thinking settings", where the effort lever lives in claude.ai and what each level costs you: https://support.claude.com/en/articles/8664678-change-the-model-effort-and-thinking-settings
• Anthropic's interactive prompt engineering tutorial (9 chapters, exercises, also as Google Sheets for non-coders): https://github.com/anthropics/prompt-eng-interactive-tutorial
• Ruben Hassid, "Prompt 4.7": https://ruben.substack.com/p/prompt-47
• For participants who want to go deeper: module 18, Prompt Engineering Deep Dive