PromptDev
Debug a performance issue
A prompt for turning performance symptoms into ranked hypotheses, each paired with the measurement that confirms it before any fix.
Last updated
Fill in the blanks
Assembled in your browser — nothing you type is stored or sent anywhere.
Your prompt
Help me debug a performance issue. Symptoms: checkout page takes 4-6s to load, but only in production; fine locally; started after last week's deploy.
Relevant stack and code: [paste the tech stack, relevant code, or stack trace here].
List the plausible causes ranked from most to least likely given the symptoms. For each hypothesis, name the specific measurement or diagnostic — a profiler trace, a query's EXPLAIN plan, a timing log, a metric to check — that would confirm or rule it out, before proposing any fix for it. Only after the ranked list, suggest a fix for the top hypothesis, labelled clearly as provisional until that measurement is actually taken. If the symptoms are equally consistent with several unrelated causes, say so instead of picking one to sound decisive.
Model-agnostic — works in Claude, ChatGPT, Gemini or any AI chat.
Why this prompt works
- →Forcing a named measurement per hypothesis before any fix stops the classic misdiagnosis loop — a confident fix for the wrong cause that costs a full deploy cycle to disprove.
- →Ranking hypotheses by likelihood, instead of listing every possible cause with equal weight, gives you an actual place to start rather than an undifferentiated checklist.
- →Labelling the suggested fix as provisional keeps the prompt honest about what it can and can't know about your system without having seen the real profiler output.