Ad Account Auditor
Audits a paid ad account against a 20-item ROAS rubric and returns a SHIP/FIX/BLOCK verdict before you launch or scale spend.
Marketing & SEOAdvanced★ 2,539⑂ 346AI score 8/10Last updated: Aug 9, 2026
What it does
Answers one question: is this paid account safe to launch or scale right now?
- Treats platform-reported ROAS as an input, never as truth — reconciles it against your own-data outcomes (deduplicated order/lead IDs from ecommerce, analytics, or CRM).
- Normalizes currency, attribution window, conversion lag, and portfolio scope before any metric comparison.
- Scores 20 typed criteria across R (measurement), O (policy/claims), A (ad safety), and S (operations) with source, date, and confidence; missing evidence stays Unknown instead of being guessed.
- Applies five veto conditions (broken instrumentation, double-counting, unapproved claims, restricted-category violations, placement safety breaches) only when positively verified.
- Outputs a SHIP / FIX / BLOCK / UNDECIDED verdict plus a prioritized fix list with owners and rerun conditions.
Safety rail: it judges only — no enabling campaigns, bid changes, budget scaling, or audience uploads without separate explicit approval.
Who it's for
- Performance marketers and growth leads managing meaningful spend on Meta, Google, or TikTok
- Agencies and freelancers who need a defensible audit report when taking over a client account
- Ecommerce operators whose dashboard ROAS doesn't match actual revenue
- Marketing managers who need evidence documentation to get a budget increase approved
(For campaign structure use campaign-architect; for creative production use ad-creative-builder.)
Examples
- Pre-scale gate: "Audit this USD account for direct response using 7-day click, 3-day lag, and a $120 CAC ceiling" → returns scored criteria, a FIX verdict, and the top three blockers.
- Attribution inflation check: supply platform conversion exports plus deduplicated order-ID exports to verify whether material double-counting triggers the R2 veto.
- Inherited account diagnostic: hand over campaign, search-term, placement, and change-history exports to surface wasted spend and unsafe placements, then draft a client-ready audit report.
· · · Install guide · · ·
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the ad/activate/ad-account-auditor folder from the GitHub repo aaron-he-zhu/aaron-marketing-skills into my ~/.claude/skills/ad-account-auditor/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/aaron-he-zhu/aaron-marketing-skills.git ~/.claude/skills/aaron-marketing-skills⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal (Terminal on macOS/Linux, Git Bash or WSL on Windows).
- Create the skills directory if needed:
mkdir -p ~/.claude/skills - This skill reads shared
references/files and a scorer script from the repo, so clone the whole repository:git clone https://github.com/aaron-he-zhu/aaron-marketing-skills.git ~/.claude/skills/aaron-marketing-skills - Verify the install:
ls ~/.claude/skills/aaron-marketing-skills/ad/activate/ad-account-auditor/SKILL.md - Restart Claude Code (or start a new session) and confirm
ad-account-auditorappears in your skill list. - Prepare your inputs: campaign / search-term / placement / change-history exports, deduplicated order or lead IDs, and the required context — currency, attribution window, conversion lag, and CAC or margin constraint.
- Prompt Claude with "Audit this ad account" plus those files and constraints. If required context is missing it will return NEEDS_INPUT/UNDECIDED, so supply everything up front.
View source on GitHub ↗License: Apache-2.0