mobile-security-mcp
An MCP server that enables AI assistants to analyze Android APK and iOS IPA files for security issues through natural language conversation, including permission auditing, secret detection, and SDK enumeration.
README
<picture> <source media="(prefers-color-scheme: dark)" srcset="docs/social-preview-dark.svg"> <img src="docs/social-preview-light.svg" alt="mobile-security-mcp" width="100%"> </picture>
<p align="center"> <a href="https://www.npmjs.com/package/mobile-security-mcp"><img src="https://img.shields.io/npm/v/mobile-security-mcp?color=0ea5e9&label=npm" alt="npm version"></a> <a href="https://www.npmjs.com/package/mobile-security-mcp"><img src="https://img.shields.io/npm/dm/mobile-security-mcp?color=0ea5e9&label=downloads" alt="npm downloads"></a> <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-10b981" alt="MIT License"></a> <img src="https://img.shields.io/badge/TypeScript-5.3-3178c6" alt="TypeScript"> <img src="https://img.shields.io/badge/MCP-compatible-6366f1" alt="MCP"> <a href="https://github.com/Serhatcck/mobile-security-mcp/actions"><img src="https://img.shields.io/github/actions/workflow/status/Serhatcck/mobile-security-mcp/ci.yml?label=CI" alt="CI"></a> </p>
<p align="center"> <a href="#features">Features</a> · <a href="#installation">Installation</a> · <a href="#tools">Tools</a> · <a href="#usage">Usage</a> · <a href="#contributing">Contributing</a> · <a href="#security">Security</a> </p>
mobile-security-mcp is an MCP (Model Context Protocol) server that gives Claude — and any MCP-compatible AI client — the ability to analyze Android APK and iOS IPA files for security issues through natural language conversation.
Security researchers, mobile pentesters, and app developers can now audit permissions, extract API endpoints, detect hardcoded secrets, inspect Firebase configuration, and enumerate third-party SDKs by simply asking Claude — no scripting required.
Features
Android
| Tool | What it does |
|---|---|
apk_manifest_analyzer |
Parses AndroidManifest.xml — flags debuggable, allowBackup, exported components, intent filters |
apk_permissions_checker |
Categorizes all permissions into dangerous vs normal with risk explanations |
android_api_extractor |
Decompiles smali bytecode to extract Retrofit HTTP endpoints and OkHttp3 fields |
android_google_services |
Extracts Firebase/GCP config from google-services.json and resources.arsc string values |
android_secrets_scanner |
Scans DEX bytecode + resources.arsc + assets for hardcoded API keys and credentials |
iOS
| Tool | What it does |
|---|---|
ios_manifest_analyzer |
Parses Info.plist — flags ATS misconfigs, URL schemes, background modes |
ios_permissions_checker |
Categorizes privacy permission declarations by HIGH / MEDIUM / LOW risk |
ios_entitlements_checker |
Extracts entitlements via codesign — flags get-task-allow, sandbox bypass, iCloud containers |
ios_binary_strings |
Extracts URLs, emails, IPs, and API key patterns from the Mach-O binary |
ios_frameworks_detector |
Lists bundled frameworks, maps ~60 known SDKs (analytics, ads, attribution, crash reporting) |
ios_google_services |
Parses GoogleService-Info.plist for full Firebase configuration |
ios_secrets_scanner |
Scans app binary + resource files for hardcoded secrets and credentials |
Shared Pattern Registry
All secret and Google service detection patterns live in a single patterns.ts — easy to extend, used by both Android and iOS scanners.
Installation
npm install -g mobile-security-mcp
Configure Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"mobile-security-mcp": {
"command": "npx",
"args": ["mobile-security-mcp"]
}
}
}
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Run from source
git clone https://github.com/Serhatcck/mobile-security-mcp.git
cd mobile-security-mcp
npm install && npm run build
{
"mcpServers": {
"mobile-security-mcp": {
"command": "node",
"args": ["/absolute/path/to/mobile-security-mcp/dist/index.js"]
}
}
}
Usage
Once configured, restart Claude Desktop and start a conversation:
"Analyze the permissions in /path/to/app.apk"
"Check this IPA for hardcoded API keys: /path/to/app.ipa"
"What Firebase services does this APK use?"
"Are there any exported components in this APK that could be an attack surface?"
"Show me all third-party SDKs in this iOS app and flag any privacy risks"
Prerequisites
Android:
apktool— required forandroid_api_extractor(brew install apktool)aapt(optional) — speeds up manifest parsing, part of Android SDK build tools
iOS (macOS only):
codesign,plutil,strings— all built into macOS, no install needed
Tools
apk_manifest_analyzer
Input: apk_path (string)
Output: Package info, security flags, components, intent filters, warnings
apk_permissions_checker
Input: apk_path (string)
Output: Dangerous permissions (highlighted) + normal permissions + risk summary
android_api_extractor
Input: apk_path OR smali_folder (string), output_format (txt|postman)
Output: Retrofit HTTP endpoints or Postman collection JSON
android_google_services
Input: apk_path (string), smali_folder (optional)
Output: Firebase project ID, API keys, database URL, storage bucket, OAuth clients
android_secrets_scanner
Input: apk_path (string), smali_folder (optional), min_length (default 8)
Output: Hardcoded credentials found in DEX + resources.arsc + assets
ios_manifest_analyzer
Input: ipa_path (string)
Output: Bundle info, ATS settings, URL schemes, background modes, warnings
ios_permissions_checker
Input: ipa_path (string)
Output: Privacy permissions grouped by HIGH/MEDIUM/LOW risk with usage descriptions
ios_entitlements_checker
Input: ipa_path (string)
Output: Entitlements extracted from binary, high-risk flags, simulator detection
ios_binary_strings
Input: ipa_path (string), filter (all|url|key|email|ip), min_length (default 6)
Output: Filtered strings from Mach-O binary
ios_frameworks_detector
Input: ipa_path (string)
Output: Bundled frameworks grouped by category with privacy risk annotations
ios_google_services
Input: ipa_path (string)
Output: Full GoogleService-Info.plist contents + pattern scan of resource files
ios_secrets_scanner
Input: ipa_path (string), min_length (default 8)
Output: Secrets found in resource files and binary, split by layer with severity
Demo

Regenerate with VHS:
brew install charmbracelet/tap/vhs && vhs docs/demo.tape
Contributing
See CONTRIBUTING.md for development setup, how to add new tools, and PR guidelines.
Security
See SECURITY.md for how to report vulnerabilities privately.
Changelog
See CHANGELOG.md.
License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。