3 ready prompts
Written for the moment this actually comes up.
Execute — do the immediate task
+I need to inspect variable values to explain a miscomputed total to Priya. Pause the running test…
Execute — do the immediate task
+I need to inspect variable values to explain a miscomputed total to Priya. Pause the running test at the failing assertion, open the locals and watches for the current frame, and capture the values of cart.items, item.price, item.quantity, and runningTotal. Copy those values into a short snippet I can paste into Slack. Ladder L1
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Improve — make it easier to accept
+Before I present values to Priya, make them easy to read: show expanded object properties, collapse…
Improve — make it easier to accept
+Before I present values to Priya, make them easy to read: show expanded object properties, collapse irrelevant system fields, convert timestamps to ISO, and display numbers with two decimals. Flag any values that are NaN, null, or zero where nonzero is expected. If any collection is long, surface the first five items and summarize the rest. Ladder L2
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Decide — diagnose the stuck moment
+The test paused and runningTotal is NaN, but I’m not sure whether a single price became NaN or an…
Decide — diagnose the stuck moment
+The runningTotal is NaN in the paused frame
The test paused and runningTotal is NaN, but I’m not sure whether a single price became NaN or an arithmetic operation did it. I cannot see earlier mutations in this frame. Should I backtrack to where runningTotal is initialized, add a watch on every place it’s written, or instrument the addition operation? What approach will most quickly reveal the root cause? Ladder L5
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Next to this one
Other code editor work people do in Visual Studio Code.
Every task here came from the work, not from a feature list — which is why the prompts name what you want done and never the button that does it. The tool changes; the work does not.
Copyright © LLOS.ai · 2026 — original pedagogy, voice, and design — all rights reserved.
The rest of the map
Same library, five ways in.