Mobile Pentest logo

Mobile Pentest

Community
hypnguyen1209
mobile-pentest

Use when pentesting an Android/iOS app — Frida 17 instrumentation, SSL-pinning & root/jailbreak bypass, Android 14/15 CA injection, exported-component/content-provider abuse, deep-link/WebView chains, biometric bypass, Flutter/React-Native RE

Overview

Publisherhypnguyen1209
Repositoryoffensive-claude
Skill namemobile-pentest
Stars
396
Forks
71
Bundled files
12
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 12 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by hypnguyen1209 on GitHub. Read the source before you install it.

Installation

Install the Mobile Pentest AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/hypnguyen1209/offensive-claude/tree/main/skills/mobile-pentest
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/hypnguyen1209/offensive-claude.git /tmp/offensive-claude
mkdir -p .claude/skills
cp -r /tmp/offensive-claude/skills/mobile-pentest .claude/skills/mobile-pentest
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Mobile Pentest in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Mobile Pentest on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Mobile Pentest is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Mobile Application Penetration Testing

When to Activate

  • Android/iOS application security assessment, bug-bounty mobile triage, or app-store reconnaissance
  • Need to intercept TLS traffic (SSL/cert pinning, Android 14/15 Conscrypt-APEX trust store, Flutter/RN stacks)
  • Bypass root/jailbreak or biometric-gating controls during dynamic analysis
  • Enumerate and exploit exported components, content providers, deep links, and WebViews
  • Extract secrets from insecure storage (SharedPrefs, SQLite, Keychain, Keystore) and reverse hybrid apps

Technique Map

TechniqueATT&CKCWEReferenceScript
Lab build + Frida 17 server/gadgetT1635CWE-1188references/environment-interception.md-
Android 14/15 Conscrypt-APEX CA injectionT1521.001CWE-295references/environment-interception.mdscripts/android_ca_inject.sh
SSL/cert-pinning bypass (Java TM/OkHttp/native)T1521.001CWE-295references/environment-interception.mdscripts/universal_unpin.js
Root detection bypass (RootBeer/native stat)T1633.001CWE-693references/environment-interception.mdscripts/universal_unpin.js
Exported activity/service/receiver abuseT1626.001CWE-926references/android-component-attacks.mdscripts/manifest_attack_surface.py
Content-provider SQLi / path traversal (CVE-2025-48609)T1409CWE-22, CWE-89references/android-component-attacks.mdscripts/component_fuzz.sh
Task hijacking / StrandHogg / TapTrap (USENIX '25)T1517CWE-1021references/android-component-attacks.mdscripts/manifest_attack_surface.py
Deep-link / intent-redirect / scheme hijackT1635, T1577CWE-939references/webview-deeplink-exploitation.mdscripts/component_fuzz.sh
WebView JS-interface RCE + file:// theftT1577CWE-749references/webview-deeplink-exploitation.mdscripts/component_fuzz.sh
OAuth custom-scheme callback interceptionT1635CWE-940references/webview-deeplink-exploitation.md-
Insecure storage (SharedPrefs/SQLite/external)T1409CWE-312references/insecure-storage-crypto.mdscripts/manifest_attack_surface.py
Keystore/Keychain misuse + dumpingT1634CWE-522references/insecure-storage-crypto.mdscripts/ios_bypass_suite.js
Biometric bypass (BiometricPrompt/LAContext)T1634CWE-287references/insecure-storage-crypto.mdscripts/ios_bypass_suite.js
iOS jailbreak + JB-detection bypassT1635CWE-693references/ios-offensive.mdscripts/ios_bypass_suite.js
IPA decrypt / class-dump / URL-scheme abuseT1409, T1635CWE-200references/ios-offensive.mdscripts/ios_bypass_suite.js
Flutter RE / reFlutter pinning bypassT1521.001CWE-295references/crossplatform-re-instrumentation.mdscripts/hermes_triage.py
React Native Hermes bytecode decompileT1640CWE-656references/crossplatform-re-instrumentation.mdscripts/hermes_triage.py

Quick Start

bash
# ---- ANDROID ----
# 0. Pull + statically triage the APK (manifest, secrets, exported surface, framework ID)
adb shell pm path com.target.app                              # locate split APKs
adb pull /data/app/.../base.apk .
python3 scripts/manifest_attack_surface.py base.apk -o surface.json
jadx -d src base.apk &  apktool d base.apk -o decoded

# 1. Frida 17: match server to host tools; push + run
frida --version                                              # e.g. 17.x  -> use matching server
adb push frida-server-17.x-android-arm64 /data/local/tmp/frida-server
adb shell "su -c 'chmod 755 /data/local/tmp/frida-server && /data/local/tmp/frida-server &'"

# 2. Trust Burp CA on Android 14/15 (APEX is immutable -> Zygote namespace bind-mount)
bash scripts/android_ca_inject.sh 9a5ba575.0 cacert.pem      # see reference for cert hashing

# 3. Spawn target with universal unpinning + root-detection bypass
frida -U -f com.target.app -l scripts/universal_unpin.js --no-pause

# 4. Hit the exported attack surface from surface.json
bash scripts/component_fuzz.sh com.target.app surface.json

# ---- iOS ----
frida-ios-dump -o app.ipa com.target.app                     # decrypt + pull (jailbroken)
frida -U -f com.target.app -l scripts/ios_bypass_suite.js --no-pause   # JB + pinning + biometric + keychain

# ---- HYBRID ----
file decoded/assets/index.android.bundle                     # "Hermes JavaScript bytecode" => RN
python3 scripts/hermes_triage.py base.apk                    # detect Flutter/RN, drive reFlutter/hermes-dec

OPSEC & Detection (summary)

TechniqueTelemetry / IOCDetection (Sigma / app-side)OPSEC note
Frida instrumentationfrida-server/gadget ports (27042), re.frida.server, suspicious maps regions, named pipes linjectorApp scans /proc/self/maps for frida, checks D-Bus port, thread gum-js-loop; Play IntegrityUse frida-gadget renamed lib, custom port frida-server -l 0.0.0.0:1337, magisk-hide / Shamiko
CA injection (APEX)New trust anchor in process trust store; cert CN mismatch on pinned hostsNetwork Security Config <trust-anchors> excludes user store; pinning catches itMount lives only in Zygote ns, vanishes on Zygote crash — re-inject; nothing written to /system
SSL-pinning bypassTLS handshake to proxy IP; cert chain not app-pinned certApp-side pin failure callbacks fire (if logged); telemetry SDK sees proxy certHook before first request; for Flutter prefer reFlutter patch over runtime to avoid crash loops
Root/JB bypassgetprop ro.debuggable, su binaries, magisk paths queriedRootBeer/iXGuard SDK reports; SafetyNet/Play Integrity attestation server-sideBypass client checks only; server-side attestation (Play Integrity / DeviceCheck) is unaffected
Exported component abuseam start/startservice/broadcast from adb; foreign UID intentApp logs unexpected caller UID; Binder.getCallingUid() checksUse on-device malicious app for realism; adb leaves shell history
Content-provider traversalcontent query with ../; openFile on out-of-dir pathFileProvider canonical-path check; CVE-2025-48609 patched Mar 2026 SPLURL-encode ..%2f to dodge naive filters; read-only first
TapTrap / task hijackTransparent activity transition, taskAffinity overlap, animationScale abuseTapTrap fixed Dec 2025 SPL; check targetSdk, taskAffinity=""Zero-permission; works <Dec-2025 patch; user study: 100% missed at least one variant
Keychain/Keystore dumpkeychain_dumper, frida memory scrape, SecItemCopyMatching hooksKeychain items with .biometryCurrentSet resist hooking; SE-backed keys unextractableJB device dumps keychain plaintext regardless of ACL; flag SE vs SW keys in report
Biometric bypassLAContext evaluatePolicy / BiometricPrompt callback hookOnly works on boolean-result pattern, NOT SecAccessControl-gated cryptoReport root cause: auth result not bound to a crypto/keychain operation

Deep Dives

  • references/environment-interception.md — Rooted/jailbroken lab, Genymotion/AVD, Magisk+Shamiko, Frida 17 breaking changes (bridge removal, frida-pm, frida-compile), gadget mode for non-root, Android 14/15 Conscrypt-APEX CA injection via Zygote namespace bind-mount, universal SSL-pinning bypass (Java TrustManager/OkHttp/Network Security Config/native BoringSSL), root-detection bypass.
  • references/android-component-attacks.md — Manifest attack-surface enumeration, exported activity/service/broadcast/provider exploitation, content-provider SQLi & path traversal (CVE-2025-48609 MmsProvider), intent:// redirection to non-exported components, task hijacking (StrandHogg 1.0/2.0 CVE-2020-0096) and TapTrap animation-driven tapjacking (USENIX Security '25), drozer + adb workflows.
  • references/webview-deeplink-exploitation.md — addJavascriptInterface RCE, setAllowUniversalAccessFromFileURLs/file:// local-file theft, deep-link → WebView open-redirect/XSS chains, intent:// browsable-activity pivots, OAuth custom-scheme callback hijack (RFC 8252), iOS URL-scheme & Universal Link abuse, one-click browser-to-WebView exploitation.
  • references/insecure-storage-crypto.md — Android SharedPreferences/SQLite/internal+external storage, Android Keystore misuse (non-hardware-backed keys, no setUserAuthenticationRequired), iOS Keychain ACLs & keychain_dumper, NSUserDefaults/plist leaks, biometric bypass (BiometricPrompt CryptoObject vs result-only, LAContext evaluatePolicy), hardcoded secrets & Firebase/S3 misconfig, MASVS-STORAGE mapping.
  • references/ios-offensive.md — Jailbreak tooling matrix (palera1n checkm8 A8–A11/T2 iOS 15–18.x, Dopamine A8–A16 iOS 15–16.6.1, Dopamine HideJailbreak), JB-detection bypass (Frida stat/fopen/dlopen hooks + Shadow), IPA decryption (frida-ios-dump/bagbak), class-dump/Swift demangling, entitlements & URL-scheme analysis, Frida-version pinning gotchas.
  • references/crossplatform-re-instrumentation.md — Flutter Dart-stack interception (reFlutter libflutter.so patch of ssl_crypto_x509_session_verify_cert_chain, iptables/proxydroid fallback), React Native Hermes bytecode RE (hermes-dec, hbctool patch/reassemble, hermes-decomp, CatalystInstanceImpl.loadScriptFromAssets Frida hook), native .so JNI/JNI_OnLoad analysis in Ghidra/IDA.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Mobile Pentest AI skill do?

Use when pentesting an Android/iOS app — Frida 17 instrumentation, SSL-pinning & root/jailbreak bypass, Android 14/15 CA injection, exported-component/content-provider abuse, deep-link/WebView chains, biometric bypass, Flutter/React-Native RE

Why use Mobile Pentest on TypingMind?

Because you install it once and use it with any model. Mobile Pentest is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Mobile Pentest in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/hypnguyen1209/offensive-claude/tree/main/skills/mobile-pentest. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Mobile Pentest?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Mobile Pentest?

As many as you like. As long as a model supports skills, you can use Mobile Pentest with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Mobile Pentest AI skill free?

Yes. It is published on GitHub by hypnguyen1209 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇