Performance sniff
Catch scaling problems before they hit production.
The prompt
Scan @<file> for performance issues that matter at scale — N+1 queries, needless allocations in hot paths, blocking calls in async code. Only flag things with real impact, and estimate the cost of each. Why it works
“Only flag things that matter at scale” plus a cost estimate keeps the model from micro-optimizing cold paths. You get a short list of changes worth making.
Fill in
@<file>— the code on a hot path.
Follow-ups
- “Fix the N+1 and show the before/after query count.”