@Ryrenz: Most people using coding AI just say “help me fix this bug,” then spend half an hour wrestling with it. Change the phrasing and it gets it right the first time. 1. Have it look before acting “Don't write code yet. Read the relevant files and tell me how this feature currently works; I'll confirm before you change it.” 2. Stuck bug “Don't guess. Add log…

Most people, when using programming AI, just say “Help me fix this bug,” and then spend half an hour going back and forth with it. Change your wording, and it’s right on the first try.

1. Make it look before it acts “Don’t write code yet. Read through the relevant files first, tell me how this feature currently works, and I’ll confirm before you change anything.”

2. Stuck bugs “Don’t guess. Add logging to print out the intermediate states, run it once, locate the issue from the real output, then fix it.”

3. Keep it from casually changing a bunch of unrelated stuff “Only change what’s necessary to make this test pass. Don’t touch other files, and leave any formatting alone.”

4. Taking over an unfamiliar project “Assume it’s my first day in this repo. Tell me where the entry point is, how the data flows, and which files are core. Don’t change anything yet.”

5. Let it find work on its own “Run the tests and type checks once, list all failures sorted by severity, and I’ll pick which ones you fix.”

6. Refactoring without derailing “Before refactoring, add tests to lock in the current behavior. Once the tests are green, then change the structure, and rerun after every step.”

7. Curing its chatty responses “From now on, only give conclusions and changes in your responses. Don’t restate my requirements, and don’t summarize what you did.”

8. Get it to admit when it doesn’t know “If you’re not sure about this part, just say you’re not sure—don’t give me a seemingly reasonable answer. I’d rather you ask me.”

9. Writing commit messages “Look at the git diff and the style of the last ten commits, and write this commit message in the same tone and format.”

10. Making technical decisions “Give me two options, spell out the cost of each and in what scenarios you’d regret it, and end with one recommendation and your reasoning. Don’t just list advantages.”

11. Keep long tasks on track “Before you start, write me a plan broken into steps. Pause after each step and wait for my confirmation before moving on.”

12. Make rules stick long-term “Write this agreement we just reached into the project’s AGENTS.md, and follow it from now on.”

The last one is the most valuable: what works well isn’t the prompt itself—it’s turning effective prompts into project rules, so you don’t have to say them again every time.

Qiaomu Ge’s sharing is well worth referencing:

1. Put together a set of impressive, high-value content in one domain.
2. Keep at it consistently and show up on schedule.
3. Manage your posting volume and timing well—mealtimes and weekends are good for long-form content.
4. Product posts must be personally tested; share experiences that exceeded your own expectations.
5. After reading enough good content, you’ll naturally want to think, test, and comment—just post it directly.