Error to Debug Prompt Converter
Paste an error and get a prompt that makes an AI diagnose it instead of guessing at a fix. The difference is the order: cause first, change last.
Your prompt
Remove any password, key or customer name from the error before you paste it anywhere.
Why pasting the error alone goes wrong
Paste a stack trace and ask what is wrong, and a model will confidently hand you a fix. Sometimes it is right. Often it has pattern-matched the error text to the most common cause and written a patch for a problem you do not have, and you spend an hour applying it before finding out.
It has no idea what changed on your site yesterday, whether it happens every time, or what you already ruled out. It does not ask, because nothing told it to.
What this prompt changes
- It asks the model to explain what the error actually means first, including what the file and line are telling you.
- Then the possible causes, most likely first, with what would make each more or less likely.
- Then what to check or log to tell them apart before changing anything.
- Only then, the smallest change that fixes the cause rather than the symptom.
- And if there is not enough information, to ask rather than guess.
The field people skip
What changed just before. Updates, a new plugin, a host migration, an expired certificate. In practice this is the answer more often than the stack trace is, and it is the one thing an AI cannot infer.
Before you paste anything
Strip passwords, API keys, customer names and email addresses out of the error first. A stack trace often carries a database password or a token, and once it is in a chat log you have to rotate it.
If the error is on a WordPress site and you would rather not do this at all, maintenance covers it, and a hacked site is a different procedure entirely. For writing the code rather than fixing it, the WordPress prompt builder is the other half.
Is it safe to paste an error into an AI?
Only after removing credentials and personal data. Stack traces frequently contain database passwords, tokens and customer email addresses.
Why not just ask for the fix?
Because a fix for the wrong cause costs more time than the diagnosis would have. The prompt forces the order.
What if the AI still guesses?
The prompt ends by telling it to ask for what is missing instead. If it guesses anyway, tell it what you have ruled out and it will narrow.
Does this work for JavaScript errors?
Yes, and for Node builds, Shopify and WooCommerce. Pick the closest option so the answer uses the right conventions.
Stuck on something that will not go away?
Send us the error and what you have tried. We debug WordPress and WooCommerce every week.