Score launched paid ads from real metrics and produce cycle artifacts.
npx skills add https://github.com/hungv47/meta-skills --skill evaluate-ad
Please help me install this Agent Skill. Skill: evaluate-ad Source: https://github.com/hungv47/meta-skills/tree/main/skills/marketing/evaluate-ad Command: npx skills add https://github.com/hungv47/meta-skills --skill evaluate-ad
Scores a launched paid ad (Meta, Google, TikTok, LinkedIn) from real metrics inside an existing eval loop, producing a cycle artifact and a ledger row to ground future decisions without generating new creative or altering channel strategy.
Run evaluate-ad for a given loop with a specific network and measurement window to produce the eval artifact and ledger row.
High-intent search queries and answers about installing and using this skill.
Ad evaluation works on one network per cycle to maintain audience-temp fidelity, preventing mixed audience signals from distorting your performance metrics and routing suggestions.
The ad evaluation process outputs a structured verdict, routing suggestions for downstream work, and a provenance trail linking evidence back to the source ad-copy artifact and loop context.
Yes, ad evaluation supports Meta, Google, TikTok, and LinkedIn networks, enforcing one-network-per-cycle discipline to maintain audience-temp fidelity and produce structured verdicts.
The best way to track ad performance trends is running evaluation cycles that generate a ledger row each time, maintaining provenance trails linking evidence to source ad-copy artifacts for future analysis.
Yes, ad evaluation outputs structured verdicts and routing suggestions for downstream work such as write-ad, plan-campaign, and run-pipeline to close the loop between hypothesis and execution.
Ad evaluation enforces one network per cycle to maintain audience-temp fidelity, ensuring that fatigue signals and performance guardrails remain accurate for the specific network being measured.
hungv47Score launched landing pages using analytics, experiments, and user recordings.
igamenovoerPlan typed local-close and relay loop runs for Houmao agent teams.
Atrium-HermesAggregate skill outputs and memory logs to rank convergent write signals.
davekim917Analyze dashboards, metrics, and BI artifacts for clarity and accuracy.
