ads-audit
Full multi-platform paid advertising audit with parallel subagent delegation. Analyzes Google Ads, Meta Ads, LinkedIn Ads, TikTok Ads, and Microsoft Ads accounts. Generates health score per platform and aggregate score. Use when user says "audit", "full ad check", "analyze my ads", "account health check", or "PPC audit".
90
Quality
86%
Does it follow best practices?
Impact
96%
2.82xAverage score across 3 eval scenarios
Evaluation results
92%
70%FreshBrew Coffee Co. — Google & Meta Ads Review
Output files and per-platform scoring structure
ADS-AUDIT-REPORT.md created
0%
100%
ADS-ACTION-PLAN.md created
0%
100%
ADS-QUICK-WINS.md created
0%
100%
Google category weights
0%
100%
Google weight values
0%
100%
Meta category weights
75%
100%
Meta weight values
0%
100%
Pass/warning/fail labels
0%
100%
Platform Health Score with grade
50%
100%
Priority level labels used
100%
100%
Remediation steps present
50%
100%
Platform-specific Quick Wins in report
0%
0%
Without context: $0.6316 · 4m 41s · 20 turns · 26 in / 13,172 out tokens
With context: $1.1547 · 7m 55s · 25 turns · 282 in / 24,296 out tokens
100%
58%Meridian Fitness — Three-Platform Paid Ads Assessment
Aggregate scoring formula and cross-platform analysis
Aggregate formula applied
0%
100%
Correct aggregate value
0%
100%
Correct aggregate grade
0%
100%
Per-platform grades correct
30%
100%
Budget allocation assessment
87%
100%
Tracking consistency check
87%
100%
Creative consistency check
62%
100%
Attribution overlap check
12%
100%
Platform prioritization
37%
100%
Kill list present
50%
100%
Scaling opportunities present
85%
100%
Budget reallocation present
85%
100%
Without context: $0.4690 · 3m 44s · 11 turns · 16 in / 10,939 out tokens
With context: $0.8521 · 5m 17s · 24 turns · 28 in / 16,267 out tokens
96%
56%Stackr SaaS — Pre-Launch Ads Account Cleanup
Quick wins identification and action plan prioritization
Only Critical/High in Quick Wins
33%
100%
Only <15-min fixes in Quick Wins
66%
100%
Quick wins sorted by impact
50%
62%
Action plan Critical section first
70%
100%
Action plan ordering complete
30%
100%
Executive summary aggregate score
0%
100%
Top 5 critical issues listed
50%
87%
Top 5 quick wins in summary
12%
100%
Business type detected
37%
100%
Active platforms listed
75%
100%
Per-platform scores in summary
0%
100%
Without context: $0.7473 · 5m 24s · 21 turns · 28 in / 15,082 out tokens
With context: $0.8564 · 5m 18s · 25 turns · 65 in / 16,170 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.