The feeling: Things are clicking, you’re in flow, and there’s a sense that you’re not actually done, there’s a smart next move sitting right here and you don’t want to lose the thread finding it.
This skill captures that. It looks at what just happened, what it unlocked, and asks: what’s the thing that compounds this? Not “which of my backlog items is highest priority”: that’s a different question. This is “I’m cooking right now, what’s the smart thing to do next?”
/qq-smart-next-move — infer from session context, surface the options
/qq-smart-next-move [context] — provide explicit context for what just finished
> Based on what we just built and the data we have, here are the moves I see:
>
> A. Wire up the observability hook. Enforcement exists but there’s no runtime evidence it’s working. Five minutes to confirm vs. hours debugging in production.
> - Data point: zero events in log after three sessions
> - Risk: low; hook either fires or it doesn’t
>
> B. Wire the scorecard into the morning digest. Invisible unless surfaced daily. Small effort, big payoff: you see it every morning without thinking about it.
> - Data point: digest is checked daily; dashboard is only checked when something feels wrong
> - Risk: schema change could break existing sections
>
> Recommendation: A + B together (15 min each).
>
> The question I’d want answered first: Is the guard actually blocking anything?
1. What did we just build? Specific deliverables, surprises, assumptions validated or broken
2. What did this unlock? What’s trivially easy now that was hard before?
3. What risk did we just create? Maintenance burden, dependencies, debt introduced
4. What compounds this? The move that makes what we just built more valuable, right now
5. Does this move the needle? Revenue proximity, client value, building for the business vs. building because it’s interesting
Part of Operator Skills, a collection of skills for running a real Claude Code setup.