Using the book with AI assistants

The book publishes two plain-text files for AI assistants and search tools. They follow the llms.txt convention, so many tools find them on their own.

File	What it holds	Use it when
llms.txt	A short guide: what the book covers, which chapter answers which topic, key vocabulary, and a link to every lesson	an assistant needs to find the right lesson
llms-full.txt	The full text and code of every lesson in one file	an assistant needs the actual explanation and code

Open llms.txt Open llms-full.txt

The repository also contains a docs.json file that lets Context7 index the lessons chapter by chapter, so coding assistants that use Context7 can look up a lesson while you work.

Ask from any page

Every page has a chat button for questions about the book, powered by Context7. Its hint text names the lesson, chapter, or section you are reading, and it takes the colour of the chapter. In the Reader theme panel, under “Chat button”, move it to any corner or turn it off. While it is off, the page makes no request to context7.com.

Getting accurate answers

An assistant answers best when it can see the lesson the question is about. Paste the lesson’s link or text into the conversation, or point the assistant at llms-full.txt, rather than relying on what it remembers. This prompt does both: fill in the two parts in square brackets and paste it.

PromptAsk for help with one lesson

I am reading the lesson “[lesson title]” in Game Hacking Academy. Its full text is in the book’s llms-full.txt file, which I will paste or attach. Using only that lesson, explain [the part I am stuck on] in plain language, with one small worked example. Do not give me memory addresses, offsets, or byte patterns for any game build. If the lesson does not contain a value, say so instead of guessing.

Addresses, offsets, and byte patterns belong to one exact build of one game. An assistant cannot know them for a build it has not seen, and it may still produce numbers that look plausible. Treat any such number as unverified until a debugger, a scanner, or the game’s own files confirm it.

The book’s labs are small, deterministic programs. When an assistant suggests a change to one, run the lab’s compile and test commands from the rust-labs README before trusting the result.