The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+I need to inspect config/cache.go as it was in commit 8f3b2a4. Show the file contents at that…
Execute — do the immediate task
+I need to inspect config/cache.go as it was in commit 8f3b2a4. Show the file contents at that commit, note the function changes compared to HEAD, and confirm whether the timeout constant is 30 or 20 seconds in that version.
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 forward the snapshot of config/cache.go at commit 8f3b2a4 to the ops lead, make it easy…
Improve — make it easier to accept
+Before I forward the snapshot of config/cache.go at commit 8f3b2a4 to the ops lead, make it easy for them: show only the function bodies that changed, highlight lines where the timeout or buffer sizes differ from main, and add a one-line explanation of why the change matters to deployment stability.
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
+Ops claims we increased the cache timeout in the last commit, but I’m not sure whether that change…
Decide — diagnose the stuck moment
+Ops asked if the timeout was increased but I can’t tell from the current file
Ops claims we increased the cache timeout in the last commit, but I’m not sure whether that change landed in 8f3b2a4 or later. I can pull the file at that commit but don’t know what diff to trust. What’s the fastest way to verify the exact value and who should I ping on Slack if the value differs from our deployment manifest?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+During incidents we repeatedly spend time confirming which file version is deployed and which…
Become — change the pattern
+We waste time reconciling file versions during incidents
During incidents we repeatedly spend time confirming which file version is deployed and which commit introduced a config change. How should we change our workflow so engineers can answer "what is running" in under five minutes? Recommend one concrete habit, one small tool or script, and one guardrail to adopt this quarter.
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 hosting work people do in GitHub.
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.