- Home
- Skills
- Mobile Development
- App Compliance Audit: AdMob Consent, COPPA, ATT, Privacy
More screenshots
Works with the AI Agents you already use
App Compliance Audit: AdMob Consent, COPPA, ATT, Privacy
One command for Flutter, React Native and Expo apps that finds the ads, consent, tracking and privacy faults that pass every build, and drafts your privacy label.
$15
App Compliance Audit: AdMob Consent, COPPA, ATT, Privacy
Example session with this skill installed
Is this app compliant before I publish the AdMob European message?
- Read your context and instructions
- Compiled the app compliance audit
What the skill printed
my-app (flutter)
warning [PRIV-CRASH]: a crash reporter is in the build but Crash Data is not declared in PrivacyInfo.xcprivacy
error [ADS-REVOKE]: nothing opens the privacy options form. AdMob's European message tells users the app has a link to change consent
error [KIDS-ADID]: child-directed ads, but the AD_ID permission is not removed (Families policy). Add it with tools:node="remove"
2 errors, 1 warnings.
Connects securely to your tools. The creator never sees your data.
About this skill
An app can request AdMob consent and never offer the "change your choice" link that Google's European message tells users is in the app, or send child-directed ads while still shipping the Android advertising-id permission, and no build or test will tell you. These come back weeks later as an App Review rejection, a Play policy warning, or ads booked to the wrong app.
audit.py compares what the code does with what it declares, in Flutter, React Native and Expo apps (the project type is read from pubspec.yaml or package.json):
- ADS-CONSENT: an ads SDK, but consent is never requested or the form never loaded.
- ADS-REVOKE: nothing opens the privacy options form.
- ADS-REVOKE-GATE: the privacy options row is never gated on the requirement status, so it is a dead row outside the EEA, UK and Switzerland.
- ADS-APPID: no AdMob app id, or Google's sample id, which only serves test ads.
- ADS-SHARED: in workspace mode, two apps with the same AdMob app id (a fork's leftover).
- KIDS-ATT, KIDS-ADID, KIDS-RATING, KIDS-TRACKING: child-directed ads that still allow the tracking prompt, keep the AD_ID permission, have no maximum ad rating, or a privacy manifest that says the app tracks.
- ATT-MISSING, ATT-LAUNCH: a usage string with no request, or a tracking prompt asked at launch.
- PRIV-MANIFEST, PRIV-CRASH: no iOS privacy manifest, or a crash reporter whose Crash Data is not declared.
--label drafts App Store privacy label and Play data safety answers from the dependencies (ads, Firebase Analytics, Crashlytics, auth, Firestore, purchases, photo and location plugins), with the steps to check the draft against the code. --workspace runs every app in a folder, one line each.
The SKILL.md gives the fix for each check and the rules for children and families: child-directed ad settings, the Kids Category as a one-way door, and how Play's target audience brings in the Families rules.
This is a checklist that produces evidence. It is not legal advice.
What you need: Python 3.9+. No packages to install.
Sold on its own; also runs inside the store-preflight skill when both are installed.
I build small apps on my own and have a shelf of them on the App Store and Google Play. The first run of this script found consent and advertising-id faults in apps of mine that had already passed review.
- It reads source and config files. KIDS-ADID checks the app's own manifest; the merged manifest from a build is where to confirm the fix.
- The label draft maps known dependencies only. Your own API, support forms and user content that leaves the device are added by hand.
- It shows that code and declarations agree. Whether the answers fit your audience is your call, and anything legal is a lawyer's.
- AdMob is the ads SDK it knows. Other ad networks are not checked.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Permissions
File Scopes
- Tools: Read, Bash (python3); Edit and Write when you ask Claude to apply a fix - Files: reads the app folder (or every app in a workspace folder); the script writes nothing - Network: none