ads

Plan, optimize, and audit paid advertising campaigns across Google, Meta, LinkedIn, TikTok, and X.

Updated Sep 6, 2026
One-click install
npx skills add https://github.com/oxie/prime-agent-skills --skill ads-oxie
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ads
Source: https://github.com/oxie/prime-agent-skills/tree/main/marketingskills/skills/ads
Command: npx skills add https://github.com/oxie/prime-agent-skills --skill ads-oxie

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Running paid ads without a disciplined framework leads to wasted budget, premature campaign kills, and misleading metrics. This Skill provides structured playbooks for campaign strategy, audience targeting, bidding, retargeting, and account audits across major ad platforms, with guardrails that prevent common AI-audit mistakes like inventing negative keywords or summing conversions across attribution windows. ## Core Features & Use Cases - Platform Playbooks: Decision systems for Meta (Andromeda-era creative strategy), Google Search, LinkedIn B2B, and ABM, with kill rules, breakeven CPL/CPC math, and scaling frameworks. - Audit Guardrails: Four-state pass/fail/unknown/NA scoring that separates account health from evidence coverage, plus hard stops against unsafe recommendations on live accounts. - Unit Economics: Payback-period analysis (CAC/ARPU per plan) replacing misleading blended LTV:CAC ratios for channel affordability decisions. - Use Case: A B2B SaaS team with $15k/month budget asks where to advertise. The Skill checks product-marketing context, applies the platform selection guide, recommends LinkedIn and Google Search with budget splits, and defines naming conventions, success metrics, and a scaling plan. ## Quick Start Ask the agent to plan a paid advertising strategy for your product, stating your budget, target audience, and conversion goal.

Frequently Asked Questions about ads

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I structure Google Ads campaigns for many keywords?▼

Organize campaigns by theme or intent (brand, competitor, features, pain points) with tightly themed ad groups of 5-15 keywords each. Apply consistent naming conventions, a match-type strategy, and negative keyword lists derived from actual search terms reports.

Should I use broad targeting or interest stacking on Meta ads?▼

In the Andromeda era, broad targeting with specific creative typically outperforms interest stacking. Feed audience identifiers into headlines and copy as identity-trigger keywords, and let the algorithm match creative variants to segments.

When should I kill an underperforming ad?▼

Pause new ads after they spend 2-3x target CPL with zero conversions, and pause mature ads running 1.5-2x over target CPL. Never pause on a fixed CPA threshold without checking sample size, conversion lag, and learning-phase status first.

Can I add Meta and Google conversion counts together for a total?▼

No. Meta's 7-day click and Google's 30-day windows use different attribution logic, so summing them double-counts conversions. Report them side by side and use a neutral source like GA4 or CRM data for a blended view.

Is LTV:CAC a good metric for deciding ad channel budget?▼

Blended LTV:CAC hides per-plan variance and can justify unprofitable spend. Use Payback Period (CAC divided by per-plan ARPU) instead, targeting a 3-12 month payback band before scaling a channel.

What are the limitations of AI-driven ad account audits?▼

Audits are only as good as evidence coverage; below 60% verified checks, no health score should be given. Unknowns never count as failures, and recommendations on live accounts must be draft plans with rollback steps, not applied changes.