From google-ads
Evaluate whether Smart Bidding is layered with manual guardrails (scripts, rules, monitoring).
npx claudepluginhub trueclicks/claude-plugins --plugin google-adsThis skill is limited to using the following tools:
Assess whether Google automation (Smart Bidding, PMax, broad match) is supplemented with appropriate guardrails. Pure automation without oversight leads to runaway spend, missed opportunities, or gradual performance drift. Layered control provides the best balance of AI efficiency and strategic governance.
Guides Payload CMS config (payload.config.ts), collections, fields, hooks, access control, APIs. Debugs validation errors, security, relationships, queries, transactions, hook behavior.
Builds scalable data pipelines, modern data warehouses, and real-time streaming architectures using Spark, dbt, Airflow, Kafka, and cloud platforms like Snowflake, BigQuery.
Builds production Apache Airflow DAGs with best practices for operators, sensors, testing, and deployment. For data pipelines, workflow orchestration, and batch job scheduling.
Assess whether Google automation (Smart Bidding, PMax, broad match) is supplemented with appropriate guardrails. Pure automation without oversight leads to runaway spend, missed opportunities, or gradual performance drift. Layered control provides the best balance of AI efficiency and strategic governance.
Data Source: Standard + Account Configuration Review
Standard Data:
data/account/campaigns/*/campaign.md - Bid strategies, automation settingsdata/account/bidding_strategies.md - Portfolio strategiesAdditional Review Required:
Reference GAQL:
SELECT
campaign.name,
campaign.bidding_strategy_type,
campaign.status
FROM campaign
WHERE campaign.status = 'ENABLED'
Use /google-ads:get-custom if you need to check specific automation configurations.
| Condition | Severity |
|---|---|
| Smart Bidding with no budget protection rules | Critical |
| No anomaly detection (spend spike/drop alerts) | Critical |
| No script for search term/n-gram automation | Warning |
| Automated rules not running (check execution logs) | Warning |
| Weekly human review not documented | Warning |
| Scripts duplicating functionality | Info |
Short format (default):
## Automation Layering Audit
**Account:** [Name] | **Automation coverage:** [X]% | **Guardrail gaps:** [Y]
### Critical ([Count])
- **Budget protection**: No rules configured -> Add spend alert at 80% daily budget
- **Anomaly detection**: Missing -> Configure CPA spike alerts
### Warnings ([Count])
- **Search term automation**: No n-gram script -> Deploy to catch waste between reviews
### Automation Inventory
| Layer | Coverage | Status |
|-------|----------|--------|
| Google Auto | X% | Good/Needs work |
| Rules/Alerts | X% | Good/Needs work |
| Scripts | X% | Good/Needs work |
| Human Review | X% | Good/Needs work |
### Recommendations
1. Deploy [script/rule] for [purpose]
Detailed adds: