- Home
- Skills
- Legal & Compliance
- App Privacy Pack: Store Answers and Policy Draft
More screenshots
Works with the AI Agents you already use
App Privacy Pack: Store Answers and Policy Draft
The agent drafts the store form answers and a privacy policy, for you to review and confirm before you submit or publish anything.
$12
App Privacy Pack: Store Answers and Policy Draft
Example session with this skill installed
Use app-privacy-pack on this repo. It is an Expo run tracker and we ship to both stores next week. My notes about the backend are in owner-answers.md.
- Read your context and instructions
- Compiled the app privacy
Summary
- Proposed as collected (11 data types): Name, Email Address, Health, Fitness, Precise Location, User ID, Device ID, Purchase History, Product Interaction, Crash Data, Other Diagnostic Data.
- Waiting on an owner answer (4 data types): Coarse Location (Q3), Search History (Q3), Other User Content (Q2), Performance Data (Q2).
- Used for tracking: none proposed. This rests on the owner's statement that there are no ads, on the absence of a tracking prompt, and on the app's own manifest (E044).
- Privacy policy URL to enter in App Store Connect: [NEEDS: public URL of the published policy]
Q2: What does Sentry receive besides the error itself?
Connects securely to your tools. The creator never sees your data.
About this skill
English
For mobile app developers and studios who have to fill in Apple's App Privacy details and Google Play's Data safety form, and publish a privacy policy.
Point your agent at the app's project folder. The agent drafts the store form answers and a privacy policy, for you to review and confirm before you submit or publish anything.
A bundled script scans the project's manifests and records every SDK, permission, usage description and privacy-manifest entry with its file and line. The agent turns that evidence, plus what you tell it, into the drafts. What the files cannot show becomes a question for you.
Drafts, not an audit. Check every answer against what your app really does at run time, your backend, your SDK settings, your partners and the current store forms before you submit. This is not a privacy audit or a compliance check.
What the agent writes each run
evidence.jsonanddata-inventory.csv: what was found, where, and the data types it usually means.apple-privacy-answers.md: a draft answer for each of Apple's data types: collected, linked to the user, used for tracking, purposes, and the evidence.play-data-safety-answers.md: a draft answer for each Data safety row: collected, shared, required or optional, purposes, and the evidence.privacy-policy-draft.mdand.html: a plain-language policy draft in 14 sections. The HTML is one self-contained page.owner-questions.md: the numbered questions only you can answer, each with the rows it decides.notes.md: what was scanned, what the scan cannot see, and whether the two forms and the policy agree with each other.
What it reads
package.json, pubspec.yaml, Podfile, Podfile.lock, build.gradle, AndroidManifest.xml, Info.plist, PrivacyInfo.xcprivacy, app.json. It works with React Native, Expo, Flutter and native iOS and Android projects. The catalog maps 38 SDKs and device capabilities to data types and recognises 28 more by name. Anything else is listed as uncatalogued for you to review.
Evidence, then questions
The skill instructs the agent to cite a file and line or your own statement for each answer, and to leave a row open with a question number when neither settles it. A check command confirms that the two forms, the policy and the evidence agree with each other.
Honest about its limits
These are evidence-based drafts. This is not legal advice, and it makes no promise of compliance or of acceptance by either store. You, the developer, are responsible for complete, accurate and current declarations, including what third-party code in your app does. Manifests and SDK lists cannot establish what is collected at run time or on your server, whether data is linked to identity, retention, consent, or every partner's practices; the pack tells you where it relies on your statement. The scanner reads manifests, not source code, so it misses hand-added binaries and anything configured in code. The SDK catalog and the form structure were written from vendor and store documentation and can be out of date; confirm each entry against the current store form. check shows the drafts agree with each other, not that they are true. No native Swift or Kotlin project was run end to end. The Arabic output had no legal review.
Arabic
On request the drafts are written in Arabic, with a right-to-left HTML policy page.
In the download: SKILL.md, standard-library Python scripts, references (store form structure, decision rules, policy writing, Arabic notes), templates, and a worked example you can read before you buy: a fictional Expo fitness app, where the scan found 67 evidence items and the pack leaves 11 questions for the owner.
Tried on two cases (an Expo / React Native fitness app in English; a Flutter delivery app in Arabic). The scripts have 37 unit tests.
بالعربي
لمطوّري تطبيقات الجوال الذين عليهم تعبئة نموذج الخصوصية في متجر Apple ونموذج أمان البيانات في Google Play ونشر سياسة خصوصية.
وجّه وكيلك إلى مجلد مشروع التطبيق. يكتب لك مسودة إجابات النموذجين ومسودة سياسة الخصوصية، لتراجعها وتؤكدها قبل أي إرسال أو نشر.
ماذا يكتب الوكيل
- ملف
apple-privacy-answers.mdوملفplay-data-safety-answers.md: مسودة إجابة لكل نوع بيانات مع الدليل عليها. - ملف
privacy-policy-draft: مسودة سياسة بلغة واضحة. - ملف
owner-questions.md: الأسئلة التي لا يجيب عنها إلا أنت. - ملف
data-inventory.csv: كل مكتبة وكل صلاحية وُجدت، مع الملف والسطر.
حدودها
هذه مسودات مبنية على الأدلة وليست استشارة قانونية، ولا تضمن الامتثال ولا قبول المتجر. ملفات المشروع لا تُظهر ما يجمعه الخادم ولا ما يحدث وقت التشغيل، لذلك تعتمد المسودة على ما تقوله أنت في هذه النقاط وتذكر ذلك. أنت المسؤول عن صحة ما تصرّح به. يمكن طلب المسودات بالعربية؛ جُرّب ذلك على حالة اختبار واحدة ولم تُراجع قانونيًا.
These are evidence-based drafts, not legal advice, with no promise of compliance or of acceptance by either store. You are responsible for complete, accurate and current declarations, including what third-party code in your app does. Manifests and SDK lists cannot establish run-time or server-side collection, linkage to identity, retention, consent or every partner's practices. The scanner reads manifests, not source code, and misses hand-added binaries and anything configured in code. The SDK catalog and form structure were written from vendor and store documentation and can be out of date; confirm against the current store form. The check command shows the drafts agree with each other, not that they are true. No native Swift or Kotlin project was run end to end. The Arabic output had no legal review. Tried on two cases. مسودات للمراجعة وليست استشارة قانونية.
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 2 days ago
Requires
Permissions
Reads your app's manifest files (package.json, pubspec.yaml, Podfile, Podfile.lock, build.gradle, AndroidManifest.xml, Info.plist, PrivacyInfo.xcprivacy, app.json) and any notes you point it to. Writes evidence.json, data-inventory.csv, the two store answer drafts, the policy draft (Markdown and HTML), owner-questions.md and notes.md into the output folder you name. Runs its bundled Python scripts locally; standard library only, no network calls. يقرأ ملفات المشروع ويكتب المسودات داخل مجلد تحدده أنت فقط.