agentleFS
TheMattBermanTheMattBerman/meta-ads-kit/skills/ad-creative-monitor/SKILL.md

ad-creative-monitor

Skill331 starsChanged 7 months agoMITView on GitHub

Track creative performance over time and detect fatigue before it kills ROAS. Monitors CTR decay, frequency creep, and CPC inflation at the ad level.

The file

Files
SKILL.md

Ad Creative Monitor — Catch Fatigue Early

Creative fatigue is the silent killer of ad accounts. CTR drops 0.1% per day, frequency ticks up, CPC quietly inflates — and by the time you notice in Ads Manager, you've burned through budget.

This skill watches for those signals daily and flags creatives that need rotation.


How It Works

Fatigue Signals (ranked by severity)

SignalThresholdSeverity
CTR dropping 3+ days in a row>20% decline from peak🔴 Critical
Frequency above 3.5Audience seeing ad too often🟡 Warning
CPC rising 3+ days in a row>15% increase from baseline🟡 Warning
Impressions decliningAd losing delivery🟠 Monitor

The Check

# Run fatigue check (mock/read-only via local adapter)
./run.sh fatigue

Reports

Fatigue Check

Daily scan of all active ads for fatigue signals.

Tell me: "Check for creative fatigue"
Or: "Are any ads getting stale?"

Creative Leaderboard

Rank all active creatives by efficiency (CTR × spend volume).

Tell me: "Rank my creatives"
Or: "Which creatives are strongest?"

Rotation Recommendations

Based on fatigue signals, recommend which creatives to pause and when new ones are needed.

Tell me: "What needs to be rotated?"
Or: "Which ads need fresh creative?"

Invocation

  1. Pull ad-level insights with daily time increment via ./scripts/meta-kit.sh
  2. Calculate day-over-day CTR, CPC, and frequency trends
  3. Flag any ad showing fatigue signals
  4. Compare against benchmarks in ad-config.json
  5. Present findings with clear severity ratings
  6. Recommend rotation schedule if fatigue detected
  7. Log findings to workspace/brand/learnings.md

Writes

FileWhat it contains
workspace/brand/learnings.mdFatigue patterns, creative lifespan data

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.