Explain problems clearly, learn tools from the docs, write better tickets
- Tech
- 12 minutes
- 3 steps
- Free
Many tech job postings now name AI tools. The useful skill is not typing questions: it is giving the tool the right context, checking what it says against the source, and keeping private code and data out of tools your company has not approved.
You leave with
A plain-language incident update, a way to learn a tool from its real documentation, and a ticket format people can act on.
Before you start: never put these into an AI tool
- Source code, API keys or customer data in a tool your company has not approved.
- Passwords, account numbers, or anything that unlocks something.
- Anything your employer says stays inside the company.
Safety: Use only the AI tools your company approves for code and internal information.
How to use the prompts
- Fill in the blanks under a prompt, then tap Copy prompt.
- Open Claude (opens in a new tab), or the AI tool your employer approves, and paste it into the message box.
- Send it, read the answer, and check it before you use it. Then come back here for the next step.
Words in this lesson
- Ticket:
- a written request or bug report in a tracking system.
Step 1 of 3Quick win
Explain a technical problem to non-technical people
Status updates go wrong when they guess. Say what you know, what you do not, and what happens next.
Prompt to copy
Turn my notes into a short update for non-technical coworkers: what happened, who was affected, what we did, what is next. Use only what I give you. If something is missing, tell me what it is instead of guessing. Do not reassure about anything my notes do not confirm. Notes: [your notes]
Fill in the blanks above, or replace each [bracket] yourself after pasting.
Worked example: spot the problem
Everything here is made up, for practice.
Their notes
App down 20 minutes. Cause: bad config change. Rolled back. Watching it.
What the AI wrote
The app was down for 20 minutes because of a configuration change. We rolled it back, and no customer data was affected.
What did the AI add? Show the answerHide the answer
The problem: Nothing in the notes says customer data was safe. That reassurance might be false, and it is the kind people repeat.
Fixed, using only the notes
The app was down for 20 minutes because of a configuration change. We rolled the change back and are watching it.
Step 2 of 3
Learn a tool from its real documentation
Paste the part of the official documentation you are working from. Ask questions about that text only.
Prompt to copy
Below is part of the official documentation for [tool]. Answer my questions using only this text, and quote the line you used. If the text does not say, tell me so instead of guessing. [the documentation] My question: [question]
Fill in the blanks above, or replace each [bracket] yourself after pasting.
Step 3 of 3
Write a ticket someone can act on
A good ticket says what happened, what you expected, and how to make it happen again.
Prompt to copy
Turn my notes into a ticket: title, what happened, what I expected, steps to make it happen again, and what I already tried. Use only my notes. List anything a developer would need that my notes do not include. Notes: [your notes]
Fill in the blanks above, or replace each [bracket] yourself after pasting.
Check before you use it
- Every claim about impact is something you confirmed.
- Answers about the tool are quoted from its documentation.
- No private code, keys or customer data went into an unapproved tool.
On your resume, only if you actually did this at work
Used [AI tool] to [task], [result you can show].
Name the tool only if you used it for real work.
Record what you did, freeSource of You is not affiliated with Anthropic. Claude is a trademark of Anthropic. Print this page for a one-sheet version.