docs(efficient-fable,efficient-frontier): add four delegation prompting practices

[remotesensei]

Summary

Add four prompting practices to the /efficient-fable and /efficient-frontier skills that make delegated subagent work more reliable.

Why

Both skills already tell an orchestrator how to write handoff packets and vet delegated work, but they leave out four practices that materially change the quality of what cheaper subagents return. In practice, subagents guess intent, over-reach scope, and report "done" without evidence — the additions close those gaps directly.

What changed

• State the "why" in every handoff: one line of operational intent (what decision the output feeds), so the subagent makes correct micro-decisions instead of guessing.
• Explicit negative scope ("don'ts"): keeps cheaper models from refactoring beyond the ask, adding features, or fixing unrelated code.
• Act once you have enough info: don't loop on planning/re-delegating, and match effort/model tier to the task.
• Make it prove it: require the command + its output, the diff, or file:line evidence before accepting "done".

efficient-frontier is the explicit mirror of efficient-fable, so both get the same edits to stay in parity. Prose-only — no frontmatter or structure changes.

How to test

npm run check (with AGENT_NATIVE_FRAMEWORK_PATH=../agent-native) passes — the change touches only SKILL.md body prose, no skill structure or config.

Notes

No migrations, no breaking changes. Two skills in one PR because they are maintained as a mirrored pair; happy to split if you'd prefer one skill per PR.

[steve8708]

Thanks, @remotesensei, this is great

One thing to fix before merge: this branch still has the old efficient-frontier frontmatter description as a plain scalar, and the current validator rejects it because it contains model: delegate (: inside a plain scalar). Could you refresh from main or switch that description to the folded scalar form currently on main?

Once that is cleaned up I'm happy to merge