Meta Ads Policy Review — Development skill for Claude Code
Pre-flight Meta ad policy checker for Claude — Cleared / Fix / Grounded verdicts with live policy citations and rewrites, before Meta ever sees the ad.
How to install Meta Ads Policy Review
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open mikefutia/meta-ads-policy-review and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Meta Ads Policy Review does
Pre-flight Meta ad policy checker for Claude — Cleared / Fix / Grounded verdicts with live policy citations and rewrites, before Meta ever sees the ad.
Alternatives in Development
- Obsidian Ingest — Ingest a source into the vault — the vault rewrites itself around new knowledge 277 ★
- Nonstop — Autonomous work mode for Claude Code 263 ★
- Diagnose Underperformer — Diagnose why a specific Google Ads or Meta Ads campaign / ad set / ad is underperforming and prescribe the nex 48 ★
README
Ad Preflight 🛫
**Catch the Meta rejection before Meta does.**
Drop in your ad copy (one ad or a whole batch) and Ad Preflight checks every element against Meta's **live** Advertising Standards — fetched fresh from Meta's Transparency Center on every run, never a stale rule list. Every ad comes back with a verdict:
- ✅ Cleared for launch — no findings worth acting on
- ⚠️ Fix before launch — likely flag, with the specific fix
- ⛔ Grounded — clear violation; the exact phrase, the policy citation, and a rewrite that keeps the selling intent
Plus a dark-mode dashboard with every finding, citation, and rewrite in one place.
Why this matters
Ad rejections aren't just annoying — repeated disapprovals teach Meta's system to distrust your account, throttle delivery, and eventually restrict it. Almost every rejection is preventable: a personal-attribute callout ("struggling with acne?"), a health claim worded one notch too hard, a before/after implication. This is the pre-flight check that catches them while they're still free to fix.
What you need
- Claude (Desktop, claude.ai, or Claude Code)
- Your ad copy — pasted into chat or dropped in as a file
**No API keys. No Meta login. No installs.** All scripts are Python 3 standard library only.
Quick start
- Install the skill:
- Claude Desktop / claude.ai: Settings → Capabilities → Skills → upload the skill zip
- Claude Code: clone this repo into your skills folder:
git clone https://github.com/mikefutia/meta-ads-policy-review.git ~/.claude/skills/ad-preflight
- Start a new chat and say: "run ad preflight"
- Paste your ads (or try
sample-ads.mdfor an instant demo). - Answer two questions (what's the product; Special Ad Category if any).
- Get your verdicts + dashboard.
What it will not do
- Replace Meta's actual review — Meta is the final gate, and account history affects it in ways a pre-flight can't see. Treat the verdict as the floor
Related Skills
Ask Nt
Ask your knowledge vault (~/Code/knowledge) a question — search + read the notes, answer in plain language gro
Bdc Validate Share
Run all pre-flight checks on a Databricks share before registering it with SAP BDC
Cost Report
Pre-flight and post-flight cost visibility with dry-run, soft caps, and expensive-command warnings
Launch And Go
Launch-and-go mode. Runs the 10-item pre-flight checklist, stamps the per-project bug log, orchestrates subage
Kuma:Guard
🛡️ Kuma Guard — Pre-flight safety check against anti-patterns, drift, and circular deps
Pre Flight Interview
访谈式方案制定 Claude Code skill:执行前逐问逐答共同定案,避免按错误假设执行。Pre-flight interview skill for Claude Code — one-question-per-
Related Agents
Validator Agent
Final security, performance, and architecture audit before a branch is cleared for merge. Issues APPROVED or B
Ads Strategist
Paid advertising strategist - project-aware market research, campaign strategy, launch-readiness verdicts
Audit LinkedIn
LinkedIn Ads evidence and controls specialist. Returns schema-valid findings for Insight Tag and conversions,