Meta Ads Healthcheck

v1.1.0

[Didoo AI] Fast on-demand campaign status check — answers "are my ads working?" using Green/Yellow/Red thresholds. For daily routine monitoring, use meta-ads...

0· 71·0 current·0 all-time
Security Scan
Capability signals
Requires OAuth tokenRequires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name and description describe a Meta Ads health check and the skill only requires META_ACCESS_TOKEN and META_AD_ACCOUNT_ID, which are exactly the credentials needed to call the Meta Marketing API. There are no unrelated credentials, binaries, or config paths requested.
Instruction Scope
SKILL.md instructs the agent to fetch campaign/adset metrics for today and the past 7 days and to apply explicit Green/Yellow/Red thresholds. It does not direct the agent to read unrelated files, exfiltrate data to third-party endpoints, or access other system credentials. The instructions are scoped to monitoring (not making optimization changes) and explicitly defer recommendations to other skills.
Install Mechanism
This is an instruction-only skill with no install spec and no code files. Nothing will be written to disk as part of an install step, which minimizes risk from supply-chain installs.
Credentials
Only two environment values are required (META_ACCESS_TOKEN and META_AD_ACCOUNT_ID). The SKILL.md documents the ads_read scope, which is read-only and proportionate to the skill's function. No unrelated secrets or broad service credentials are requested.
Persistence & Privilege
always:false (default) and normal user-invocable/autonomous invocation settings. The skill does not request persistent or elevated system privileges and does not attempt to modify other skills' configurations.
Assessment
This skill appears internally consistent with its purpose. Before installing: (1) only provide a Meta access token with the minimum required ads_read permissions and prefer short-lived tokens you can rotate; (2) verify you trust the skill publisher (homepage is a marketing blog link) because the skill will be able to read your ad account data; (3) consider limiting the agent's autonomy or monitoring logs if you are uncomfortable with automated skill invocation; (4) revoke or rotate the token if you stop using the skill. If you need stronger guarantees, ask the publisher how they handle tokens and whether the skill stores/forwards data outside Meta's API.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

EnvMETA_ACCESS_TOKEN, META_AD_ACCOUNT_ID
didoo-aivk977srsqy11dw83vmy1f56ge3d84v930healthcheckvk977srsqy11dw83vmy1f56ge3d84v930latestvk972sjjtjmsgxe9twk2enm8wjx84zdm8meta-adsvk977srsqy11dw83vmy1f56ge3d84v930
71downloads
0stars
2versions
Updated 3d ago
v1.1.0
MIT-0

Required Credentials

CredentialWhere to GetUsed ForOAuth Scope
META_ACCESS_TOKENMeta Developer Console → Graph API Explorer → Generate TokenFetching campaign and adset dataads_read (read-only)
META_AD_ACCOUNT_IDAds Manager URL: adsmanager.facebook.com/act_XXXXXXXXXIdentifying which account to query

When to Use

When user wants a quick "are my ads working?" check — not a deep diagnostic, just a status update. Use before a meeting, or any time something feels off.

meta-ads-healthcheckmeta-ads-daily-pulse
Primary question"Is this campaign healthy or not?""Did anything change vs. last week?"
Comparison basisFixed Green/Yellow/Red thresholdsSame day of prior week (WoW)
Best forOn-demand check when something feels wrong; meeting prepDaily routine, morning scan
Output styleCampaign-by-campaign status reportChange alerts ranked by revenue impact

Step 1: Get the Basics

Pull the user's Meta Ads data for today and the past 7 days:

  • Campaign name, status, spend
  • Results, cost per result
  • CTR, frequency
  • Any alerts or issues (delivery low, learning limited, paused)

Use Meta Marketing API to fetch campaign and adset level data.


Step 2: Traffic Light Assessment

Score each campaign as Green / Yellow / Red:

SignalGreenYellowRed
DeliverySpending normally< 80% of budget< 50% or zero
Cost per resultAt or below target10–30% above target (and variance is normal)30%+ above target and sustained (not just a spike)
Frequency< 33–4> 4
LearningOut of learningIn learning (< 7 days)Stuck in learning
CTR> 1%0.5–1%< 0.5%

Step 3: Check for Normal Performance Fluctuations

Before flagging a campaign as Yellow or Red, confirm it's not normal fluctuation.

Normal fluctuation — monitor only:

  • CPA is bouncing around day-to-day (up to ±30%) but the rolling average is stable — this is normal Meta Ads volatility, not a problem
  • Weekend vs. weekday differences
  • Gradual changes over weeks
  • Variation while in Learning Phase

Concerning — investigate today:

  • CPA average has shifted up and stayed there for 3+ consecutive days (not just a one-day spike) — this is a real trend, not fluctuation
  • Delivery dropping to near zero with no budget change
  • Conversion rate declining while spend increases
  • Performance degradation after a recent edit

Before diagnosing problems, ask:

  1. Is the ad set still in Learning Phase? (if yes, delay judgment — data is unstable)
  2. What's the baseline for normal variation for this campaign?
  3. Are there external factors — seasonality, competitor activity, or platform changes?
  4. Is sample size sufficient? (typically need 7+ days of data for stable ad sets)

Step 4: Surface Key Issues

For each Yellow or Red campaign, identify:

  • What is the specific problem?
  • How urgent is it? (Needs attention today / this week / monitor)
  • What likely caused it?

Creative Refresh Trigger — Frequency > 3: When frequency reaches 3 or higher, flag the campaign as "Creative Refresh Needed": "Campaign [X] has frequency at [N] — audience is seeing the same ads too often. This typically causes CTR to drop and CPL to rise. Creative refresh is indicated."

Note: When this flag appears, the appropriate next step is meta-ads-recommendation — it will prescribe the specific creative refresh action based on the full analysis context. Do not trigger meta-ads-builder directly from this skill.


Step 5: Quick Output

Overall Status: [N] campaigns running | [N] Green | [N] Yellow | [N] Red

For each Red/Yellow campaign:
- Campaign name
- Status: [Issue]
- Spend / Results / CPL
- What happened (1 sentence)
- Suggested focus area (1 sentence)

Everything else: brief healthy summary.

Priority Sort — What to Check First

When you have 10+ active campaigns:

  1. Red campaigns with spend > $50/day → Check today
  2. Yellow campaigns where CPL is 20%+ above target → Check today
  3. New campaigns (< 7 days, still in Learning) → Flag as "Learning Phase — Do Not Judge Yet"
  4. Green campaigns → Review weekly

When to Escalate

  • If the health check surfaces a real problem: suggest meta-ads-analysis for a full diagnostic
  • For daily routine monitoring: suggest meta-ads-daily-pulse instead

Skill Boundary

  • This is a quick scan, not a deep diagnostic — don't draw conclusions beyond what's clearly broken or healthy
  • Do not make optimization recommendations in this skill — that's meta-ads-recommendation's job
  • Do not re-fetch data if it was recently pulled in the same session

Comments

Loading comments...