ppc-ai-skills

Scans your agency accounts for keywords with 0 impressions in last 180 days. Auto-invoke when user says 'scan non-serving keywords', 'find dead keywords', 'keyword cleanup report', or 'non-serving keyword scan'. Outputs to Google Sheet for human review.

Build Performance Max campaigns and output Google Ads Editor-importable CSV (115 columns, UTF-16LE). Auto-invoke when user says "build pmax", "pmax build", "new pmax campaign", "pmax csv", or references a todo file with `pmax-build` in the name. Generates CSV from a campaign brief plus YouTube videos and ad copy.

One-time setup guide and scripts for getting Google Ads API access working. Auto-invoke when user says "set up Google Ads API", "generate credentials", "create google-ads.yaml", "OAuth setup", "test API connection", or hits authentication errors like "invalid_grant", "redirect_uri_mismatch", "The developer token is not approved", or a refresh token that keeps expiring every week. Prerequisite for every skill that queries or modifies Google Ads.

Portfolio-wide scan of every manager (MCC) that has access to accounts in your Google Ads tree. Built after an MCC link-fraud incident where hostile external MCCs were found linked to client accounts via compromised admin credentials. Classifies each manager link as INTERNAL (in your tree), HOSTILE (known threat), or EXTERNAL (potential exposure — no judgment). Outputs CSV by default with optional Google Sheets upload. Auto-invoke when user says 'mcc hack audit', 'audit mcc links', 'scan for external managers', 'check manager links', 'run mcc link scan', 'find external mccs', or asks who has access to their Google Ads accounts.

Generate a full Responsive Search Ad set for ONE account — 15 headlines + 4 descriptions per active ad group — from website-verified content plus live SERP competitive analysis and a website→Google-Business-Profile review fallback. Ships all seven workflow scripts. Auto-invoke when user says "create RSAs for [account]", "generate RSAs for [account]", or "build RSA copy for [account]". Outputs an import-ready Google Sheet. Read-only on Google Ads (no mutations).

42-point Google Ads account inspection (44 with the local-service preset) — discrete GREEN/YELLOW/RED checks across 14 categories (conversion tracking, pacing, impression share, quality score, search-term waste, keyword health, creative, RSA assets, account settings, PMAX config, negative keywords, placement safety, extensions, video & Demand Gen automation), scored into one overall verdict with estimated monthly waste, then triaged per rules.md. Auto-invoke when user says 'run diagnostic', 'inspect account', '40-point inspection', 'diagnostic for [account]', 'audit [account]', or 'account health check'.

Audit Google Ads accounts for creative compliance (10 checks) with intelligence integration (issue-history comparison, chronic-issue detection, daily-briefing integration). Ships both scripts (audit + briefing reader). Auto-invoke when user says "run ads checker", "creative audit", "check ads for [portfolio]", "audit ads", or "ads checker for [account/portfolio]". Outputs to a Google Sheet with a severity breakdown. Read-only — no Google Ads mutations.

Bulk-disable Demand Gen ad-level asset automation settings. Auto-invoke when user says "disable dgen automation", "fix dgen asset settings", "turn off dgen auto assets", "dgen automation fix", or "fix demand gen automation".

Investigate Google Ads accounts with significant underspending (pacing variance beyond your portfolio's tolerance). Auto-invoke when user says "investigate [account] underspending", "[account] underspending", "why is [account] underspending", or "diagnose underspend for [account]". Runs a universal investigation script, applies six diagnostic frameworks, and synthesizes a root-cause diagnosis with a budget recommendation (or explicit no-action call). Read-only — no mutations.

Mandatory verification requirements for all ad copy creation (RSAs, extensions, callouts, structured snippets). Auto-invoke when generating ad copy. Enforces website-sourced claims only — no assumptions, no templates-as-content, no unverified placeholders. Core principle — Empty > Inaccurate.

Structured hypothesis-driven investigation framework for diagnosing Google Ads performance issues. Auto-invoke when user says "investigate," "why is [account] performing poorly," "dig into," or asks to diagnose a performance issue.

MANDATORY safety system for ALL Google Ads mutations AND destructive Google Sheets writes. Auto-invoke whenever ANY agent, skill, or script attempts to modify Google Ads accounts or overwrite Google Sheets data. Enforces two-step approval to prevent accidental changes.

Account prioritization criteria for portfolio health monitoring and daily briefings. Auto-invoke when determining which accounts need investigation, prioritizing daily actions, running portfolio health checks, or deciding investigation order. Provides 5-tier priority system with portfolio-specific thresholds (Portfolio A ±5%, Portfolio B ±8%).

Client-facing report formatting standards. Auto-invoke when creating client summaries, investigation reports, monthly reviews, performance analyses, or any communication for stakeholders. Applies mandatory Background → Analysis → Conclusions framework with required "What We Looked At" data source attribution for all findings.

Audits conversion tracking health across Google Ads portfolios. Auto-invoke when user says 'conversion tracking audit', 'check conversion health', or 'audit conversions for [portfolio]'. Only analyzes accounts with spend in last 7 days. Identifies conversion actions that haven't fired or are stale.

Query Google Ads data and save to CSV. AUTO-ACTIVATE for: search terms, campaigns, keywords, ad groups, conversions, budgets, assets, geo performance. Also triggered by account names or "get/pull/show me" + Google Ads resource.

Transforms markdown reports into professionally formatted Google Sheets for client presentation. Auto-invoke when user wants to make a report presentable, share analysis with clients, convert markdown to Sheets, or export competitive analysis. Triggers include "make this presentable", "create a Google Sheet", "export to sheets", "format for client".

Refresh RSA ad copy by scraping the business website, identifying LOW-performing assets, and generating replacement headlines and descriptions using AI. Auto-invoke when user says 'refresh RSAs for [account]', 'update RSA assets', 'rewrite RSAs', or 'replace low-performing RSA assets'. Uses Firecrawl for scraping and Claude for copy generation.

Sheet-driven Google Ads shared budget automation. Approve a budget change by adding a row (CID, shared budget ID, amount) to a Google Sheet tab; a daily GitHub Actions cron pushes it via campaignBudgets:mutate (amount only), marks the row done, and sends one consolidated Slack alert for any failed rows. Failed rows retry next run; amounts of $0 or less never reach the API. Use this skill when the user asks to "set up shared budget updater", "push budget changes from a sheet", "bulk update shared budgets on a schedule", or wants a spreadsheet-approved budget pipeline.

Analyze search queries for geographic targeting conflicts in GEO campaigns. Uses OpenAI GPT-4o to determine if queries should PASS (no conflict - safe to add as negative) or FAIL (conflict detected - do NOT negative). Auto-invoke when user mentions "geo conflict check", "analyze geo conflicts", or after running off-brand analysis.

Monitors a Google Ads MCC for monthly budget overruns and sends Slack alerts at 100% and 120% of monthly budget per account. Designed to catch hijacked accounts, runaway PMax campaigns, and pacing-sheet formula errors within 2 hours instead of 24+. Alert-only — never pauses campaigns. Runs on GitHub Actions on a 2-hour cron. Use this skill when the user asks to "set up budget guardian", "install spend alerts", monitor MCC spend, or build a tripwire against unauthorized PMax campaigns.

Comprehensive 23-element Google Ads RSA copywriting framework with distribution formulas, sentiment scoring, and the "Four Words to Value" rule. Auto-invoke when user says "generate RSA copy", "write ad copy", "RSA framework", "headlines and descriptions", or asks for ad copywriting principles. Pure protocol skill, no scripts required.

Add account-level negative keywords (Admin → Account Settings → Negative Keywords) to Google Ads accounts using the 3-step SharedSet approach. Auto-invoke when user says "add account negatives", "add account-level negatives", "set up account negative keywords for [account(s)]", or "roll the negative-list baseline to [accounts]". Supports single-account and bulk-portfolio operations.

Calculates conservative budget recommendations for Google Ads accounts based on pacing variance, impression share analysis, and performance constraints. Auto-invoke when creating budget recommendations, investigating underspending, or determining optimal budget increases. Enforces 5-10% increase limits and performance safeguards.

Query Google Ads account change history for any date range using the `change_status` resource (no 30-day limit, unlike `change_event`). Auto-invoke when user says "what changed in [account]", "change history for [account]", "what did I do in [month]", "audit account changes", or "review optimization history".

Comprehensive competitive analysis combining website intelligence, 22-attribute ad copy framework, automated screenshots via Playwright, and verification rigor. Auto-invoke when user says "analyze competitors", "competitor report", "competitive intelligence", "market positioning analysis", or "client gift". Outputs Strategic Client Gift OR tactical Ads Angle Brief.

Fair Housing Act compliance requirements for property management advertising. Auto-invoke when creating campaigns, modifying targeting settings, setting up audiences, recommending changes to location/demographic targeting, or ANY discussion of campaign targeting for property management accounts. CRITICAL LEGAL REQUIREMENT - ensures all Google Ads accounts comply with federal housing discrimination laws.

Cross-referencing framework for comparing GA4 behavioral data with Google Ads campaign settings to identify discrepancies and configuration gaps. Auto-invoke when cross-analyzing GA4 and Ads data, investigating discrepancies, or verifying campaign configuration. Uses "Hypothesis → Verification → Finding" methodology.

Collect and structure GA4 + Google Ads data for cross-platform analysis. Reusable data collection engine for lead quality investigations, conversion analysis, and audience insights. Auto-invoke when user asks "what landing pages are converting from [campaign]", "show me GA4 data for [campaign]", "what user segments are converting", or wants to cross-reference GA4 with Google Ads.

Reusable GAQL query templates for Google Ads API. Auto-invoke when writing Google Ads queries, analyzing campaign data, querying performance metrics, or building custom reports.

Reference library of official Google Ads API code samples. Auto-invoke when building NEW Google Ads scripts, agents, or functionality (campaigns, mutations, reporting, account management). Helps find relevant examples in /examples/ folder before writing custom code. Improves efficiency by starting from proven patterns rather than guessing API implementation.

Impression share analysis and root cause diagnosis for Google Ads Search campaigns. Auto-invoke when investigating underspending, analyzing auction competitiveness, or diagnosing budget vs. quality issues. Interprets Search IS, Budget Lost IS, and Rank Lost IS metrics in smart bidding context.

Red flag detection frameworks for lead quality analysis across GA4 behavioral data. Auto-invoke when analyzing lead quality, identifying bot traffic, investigating no-shows/missing contact info, or reviewing conversion patterns. Provides 5 analysis frameworks (landing pages, geo, devices, time, user behavior) with severity classification.

Recommendation prioritization framework with implementation timelines for lead quality investigations. Auto-invoke when generating recommendations, prioritizing actions, or creating implementation plans. Provides 3-tier priority system (Immediate/Medium/Long-term) with structured recommendation templates.

Audit and fix Performance Max campaign asset automation settings. Opts out of Google's auto-generated headlines, descriptions, videos, and image enhancements that often produce off-brand or hallucinated output. Auto-invoke when user says "audit PMAX asset automation", "check auto-created assets", "fix PMAX automatic assets", or "are PMAX campaigns opted out".

Portfolio-specific pacing thresholds, budget tolerances, and brand caps. Auto-invoke when analyzing pacing variance, checking if accounts are on pace, investigating underspending/overspending, or discussing portfolio-specific rules (Portfolio A ±5%, Portfolio B ±8%, Portfolio C). Also loads when user mentions specific portfolios by name.

Find and replace text across RSA ads at scale, outputting to Google Sheet in Google Ads Editor format for review before manual import. Auto-invoke when user says "RSA bulk edit", "find and replace in RSA ads", "edit RSA ads at scale", or "bulk RSA changes". Supports single account, multiple CIDs, dry-run preview, and case-sensitive matching.

Classify search terms by intent. Auto-invoke when user provides search terms to classify, asks for search term analysis, or wants to identify negative keyword candidates.

Audit Google Ads accounts for bad YouTube placements (kids content, adult, gaming, non-English, spam) across PMAX, Demand Gen, and Display campaigns. Flags violations, writes to Google Sheets, then extracts channels for bulk negation.

End-to-end search query report (SQR) negative-keyword pipeline — pull search terms across your MCC, prep classification batches, run 3 independent classification passes for consensus, (optionally) check geo conflicts, write reviewable agree tabs, then upload approved PHRASE negatives to shared lists with two-step mutation safety. Includes a remove branch to un-negate mistakes. Auto-invoke when the user says "run SQR pipeline", "classify search queries", "pull search terms", "upload negatives", "sqr uploader", "remove negative", or "un-negate keyword".