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 a quick view of recent Actions workflow runs for the frontend repo so I can tell QA whether…
Execute — do the immediate task
+I need a quick view of recent Actions workflow runs for the frontend repo so I can tell QA whether the failing PR triggered the scheduled nightly job. Show the last 30 runs, their status, triggering actor, branch or tag, and duration. Highlight runs that failed or were cancelled in the past 24 hours and include links to the failed job logs.
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 share run data with the product manager, make it easy for them to spot regressions:…
Improve — make it easier to accept
+Before I share run data with the product manager, make it easy for them to spot regressions: surface runs that changed status versus the previous run, show which commits introduced failures, and list failing jobs and error summaries up front. Point out any runs that look abnormal in duration compared to the 30-day median.
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 scheduled nightly workflow shows several recent runs, but the QA lead claims there was no…
Decide — diagnose the stuck moment
+A QA lead says the nightly job didn't run but show page lists runs.
The scheduled nightly workflow shows several recent runs, but the QA lead claims there was no deployment artifact yesterday. I only see runs for the cron but I'm unsure if the job completed successfully or artifacts were published. What's the most likely cause and the clearest next check to confirm whether the deployment step produced artifacts?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Every week I get ad-hoc requests for lists of workflow runs and failure trends across five repos;…
Become — change the pattern
+Stakeholders keep asking for run exports and I rerun queries manually.
Every week I get ad-hoc requests for lists of workflow runs and failure trends across five repos; pulling them takes me hours. I manage CI observability for the org. Where am I wasting time and what habit or small automation should I adopt so product and QA can get the run info themselves without manual fiddling?
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.