understand-prompt

Why we included this project

This is a first-person, Chinese-language account of getting real work out of generative AI tools, written by a developer who shows his working rather than hand-waving at theory. He walks through image generation with Stable Diffusion, article writing with ChatGPT, and everyday coding with Copilot, with practical sections on building context, using ControlNet for fine control, and negative prompting. A set of Jupyter notebooks for translation and code generation lets you run the same techniques yourself. The repo is more of a personal knowledge notebook than polished production software, and it does contain adult images, so it is not for minors or the office. For teams new to prompt engineering, the trial-and-error detail is the value: you get the mistakes and fixes, not just the prompts.