skills/marketing/google-ads/SKILL.md
Query, audit, and optimize Google Ads campaigns. Supports two modes: (1) API mode for bulk operations with google-ads Python SDK, (2) Browser automation mode for users without API access - just attach a browser tab to ads.google.com. Use when asked to check ad performance, pause campaigns/keywords, find wasted spend, audit conversion tracking, or optimize Google Ads accounts.
npx skillsauth add pedronauck/skills google-adsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Manage Google Ads accounts via API or browser automation.
Check which mode to use:
google-ads.yaml configured or GOOGLE_ADS_* env vars# Check for API config
ls ~/.google-ads.yaml 2>/dev/null || ls google-ads.yaml 2>/dev/null
If no config found, ask: "Do you have Google Ads API credentials, or should I use browser automation?"
Requirements: User logged into ads.google.com in browser
browser tool with profile="chrome"1. Navigate to: ads.google.com/aw/campaigns
2. Set date range (top right date picker)
3. Snapshot the campaigns table
4. Parse: Campaign, Status, Budget, Cost, Conversions, Cost/Conv
1. Navigate to: ads.google.com/aw/keywords
2. Click "Add filter" → Conversions → Less than → 1
3. Click "Add filter" → Cost → Greater than → [threshold, e.g., $500]
4. Sort by Cost descending
5. Snapshot table for analysis
1. Navigate to keywords or campaigns view
2. Check boxes for items to pause
3. Click "Edit" dropdown → "Pause"
4. Confirm action
1. Navigate to desired view (campaigns, keywords, etc.)
2. Click "Download" icon (top right of table)
3. Select format (CSV recommended)
4. File downloads to user's Downloads folder
For detailed browser selectors: See references/browser-workflows.md
Requirements: Google Ads API developer token + OAuth credentials
# Verify google-ads SDK
python -c "from google.ads.googleads.client import GoogleAdsClient; print('OK')"
# Check config
cat ~/.google-ads.yaml
from google.ads.googleads.client import GoogleAdsClient
client = GoogleAdsClient.load_from_storage()
ga_service = client.get_service("GoogleAdsService")
query = """
SELECT campaign.name, campaign.status,
metrics.cost_micros, metrics.conversions,
metrics.cost_per_conversion
FROM campaign
WHERE segments.date DURING LAST_30_DAYS
ORDER BY metrics.cost_micros DESC
"""
response = ga_service.search(customer_id=CUSTOMER_ID, query=query)
query = """
SELECT ad_group_criterion.keyword.text,
campaign.name, metrics.cost_micros
FROM keyword_view
WHERE metrics.conversions = 0
AND metrics.cost_micros > 500000000
AND segments.date DURING LAST_90_DAYS
ORDER BY metrics.cost_micros DESC
"""
operations = []
for keyword_id in keywords_to_pause:
operation = client.get_type("AdGroupCriterionOperation")
operation.update.resource_name = f"customers/{customer_id}/adGroupCriteria/{ad_group_id}~{keyword_id}"
operation.update.status = client.enums.AdGroupCriterionStatusEnum.PAUSED
operations.append(operation)
service.mutate_ad_group_criteria(customer_id=customer_id, operations=operations)
For full API reference: See references/api-setup.md
Quick health check for any Google Ads account:
| Check | Browser Path | What to Look For | |-------|--------------|------------------| | Zero-conv keywords | Keywords → Filter: Conv<1, Cost>$500 | Wasted spend | | Empty ad groups | Ad Groups → Filter: Ads=0 | No creative running | | Policy violations | Campaigns → Status column | Yellow warning icons | | Optimization Score | Overview page (top right) | Below 70% = action needed | | Conversion tracking | Tools → Conversions | Inactive/no recent data |
When reporting findings, use tables:
## Campaign Performance (Last 30 Days)
| Campaign | Cost | Conv | CPA | Status |
|----------|------|------|-----|--------|
| Branded | $5K | 50 | $100| ✅ Good |
| SDK Web | $10K | 2 | $5K | ❌ Pause |
## Recommended Actions
1. **PAUSE**: SDK Web campaign ($5K CPA)
2. **INCREASE**: Branded budget (strong performer)
google-ads.yamltools
Plans real-user QA deliverables: personas, journey maps, exploratory charters, persona/journey/tour/CFR test cases, regression suites, Figma validation checks, automation intent, and user-impact bug reports. Writes artifacts under <qa-output-path>/qa/ for qa-execution to consume. Use when planning QA before execution, documenting journey-driven test strategy, marking flows that need E2E follow-up, or filing structured bug reports. Do not use for live execution, AI implementation audits, CI gate ownership, or technical integration/security/performance suites; use qa-execution or agent-output-audit instead.
development
Executes real-user QA sessions through public interfaces using personas, journeys, exploratory charters, test tours, edge-case probes, CFR checks, and browser evidence. Reads qa-report artifacts from <qa-output-path>/qa/ when present, captures issues/screenshots/reports under the same output tree, and classifies bugs by user impact. Use when validating a release candidate, migration, refactor, or user-facing change against production-like behavior. Do not use for AI implementation audits, task-status reconciliation, CI gate runs, integration/security/performance templates, or flaky-test triage; use agent-output-audit for those.
development
Transform outside-of-diff review files into properly formatted issue files for a given PR. Use when converting review files from ai-docs/reviews-pr-<PR>/outside/ into issue format in ai-docs/reviews-pr-<PR>/issues/. Automatically determines starting issue number and preserves all metadata (file path, date, status) from original review files. Don't use for inline-diff review files, non-PR review artifacts, or creating GitHub issues directly.
development
Enforce root-cause fixes over workarounds, hacks, and symptom patches in all software engineering tasks. Use when debugging issues, fixing bugs, resolving test failures, planning solutions, making architectural decisions, or reviewing code changes. Activates gate functions that detect and reject common workaround patterns such as type assertions, lint suppressions, error swallowing, timing hacks, and monkey patches. Don't use for trivial formatting changes or documentation-only edits.