evaluate-adlisted

# Ad Eval — Orchestrator <!-- BUDGET_EXCEPTION: Eval skills carry artifact-schema-as-contract (8 body sections + 8-col results row + cross-stack consumer contract) that is load-bearing and cannot move to references/. Cycle ledger discipline requires the schema be visible in the SKILL.md body. Ad-eval also surfaces the audience-temp scoping rule (one cycle = one audience-temp) and creative-fatigue signal columns. ~300 tokens over the standard cap is the legitimate cost. --> *Evaluation skill. Converts launched Meta-ad evidence into a cycle snapshot + ledger row + narrowly-scoped next action inside an existing eval loop. One cycle = one audience-temp.* **Core Question:** "Did this ad cycle, for this audience-temp, create measurable signal strong enough to keep / discard / watch / block — and what should the next strategy/execution skill know?" > Why, methodology, history: [`references/playbook.md`](references/playbook.md) [PLAYBOOK]. Capability metadata (route triggers, prerequisites, load map): [`routing.yaml`](routing.yaml). ## Critical Gates 1. **Existing eval loop required.** `program.md` + `context.md` absent → `NEEDS_CONTEXT`, recommend `/run-eval-loop`. This skill does not create loops. 2. **Measurement evidence required + primary metric decides the row.** Need ≥1 metric source, window, current value for the loop's primary metric (CTR / CPA / ROAS / conversion rate — operator-pick via `program.md`). Secondary metrics (frequency, fatigue indicators, qualitative) exp