The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+Cancel the in-progress workflow run for branch feature/payment-retry that started at 08:14 because…
Execute — do the immediate task
+Cancel the in-progress workflow run for branch feature/payment-retry that started at 08:14 because it hung on an external service call and is blocking the runner quota. Notify the PR author and the ops on-call channel that you cancelled it and add a short note 'hung on payments service' to the run's annotation.
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 cancelling, make cancellation easy to review: show which step is hanging, how long it's been…
Improve — make it easier to accept
+Before cancelling, make cancellation easy to review: show which step is hanging, how long it's been stuck versus its normal duration, which runners it's occupying and the downstream jobs waiting. Flag whether cancelling risks leaving resources allocated or partial releases, and suggest whether to cancel now or let it finish with a timeout adjustment.
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
+A build on our self-hosted runners has been stuck for 45 minutes and several developers are blocked…
Decide — diagnose the stuck moment
+A long job is blocking our self-hosted runners and devs need them.
A build on our self-hosted runners has been stuck for 45 minutes and several developers are blocked from running tests. I can cancel it but I'm concerned it might leave cloud resources running or interfere with an ongoing deploy. I don't know if the job touches external systems. What's the cautious next step to free runners without creating side effects?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Across the team we routinely cancel workflow runs whenever external APIs are flaky, then re-trigger…
Become — change the pattern
+We cancel runs repeatedly when third-party APIs are flaky.
Across the team we routinely cancel workflow runs whenever external APIs are flaky, then re-trigger them later. This pattern wastes time and delays debugging. I maintain the CI and want to reduce cancellations. What operational habit and one automation would stop the reflex to cancel while keeping developer flow?
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.