Jadx MCP Plugin
J

Jadx MCP Plugin

該項目是一個Jadx插件,通過HTTP暴露Jadx API,實現與Claude的即時交互,支持AI輔助的Android應用逆向工程分析。
2.5分
9.7K

What is Jadx MCP Plugin?

This plugin connects Jadx (an Android decompiler) with Claude AI through MCP, allowing you to use natural language to navigate and analyze decompiled Android app code. It acts as a bridge between technical reverse engineering tools and conversational AI assistance.

How does it work?

The Java plugin exposes Jadx's decompiler API over HTTP, while a Python adapter translates between MCP protocol and the plugin. This lets Claude directly query and analyze Android app code in real-time during your conversations.

When should I use it?

Ideal for: 1) Security researchers analyzing suspicious Android apps 2) Developers learning from decompiled code 3) Anyone who wants AI assistance while reverse engineering

Key Features

Class Navigation
Browse and search all decompiled Java classes in the target APK/DEX file
Source Inspection
View complete decompiled source code of any class directly in Claude
Method Analysis
Examine individual methods including their implementation code
Field Inspection
View all class fields and their properties
Advantages
Natural language interface for reverse engineering
Real-time interaction with decompiled code
No need to manually browse through thousands of classes
Combines Jadx's accuracy with Claude's analysis capabilities
Limitations
Requires both Jadx and Claude to be running
Limited to Java/Android reverse engineering
Initial setup requires technical configuration

Setup and Usage

Install the plugin
Download the plugin JAR from releases and install it in Jadx either via GUI or CLI
Configure Claude
Add the MCP server configuration to Claude's settings file as shown in the examples
Load your APK
Open your target Android file in Jadx as you normally would
Start analyzing
Ask Claude to examine the code using natural language commands

Example Use Cases

Finding Cryptographic Operations
Locate all cryptographic functions in an app to assess security
Analyzing Network Communication
Examine how an app communicates with its servers

Frequently Asked Questions

Do I need programming knowledge to use this?
Can I use this with other AI assistants?
Is there a performance impact?

Additional Resources

Jadx GitHub
The main Jadx decompiler project
MCP Protocol Documentation
Official Model Context Protocol specifications
Mobile Hacking Lab
Creators of this plugin

安裝

複製以下命令到你的Client進行配置
{
  "mcpServers": {
    "Jadx MCP Server": {
      "command": "C:\\Workset\\jadx-mcp-plugin\\venv\\Scripts\\python.exe",
      "args": ["C:\\Workset\\jadx-mcp-plugin\\fastmcp_adapter.py"]
    }
  }
}

{
  "mcpServers": {
    "Jadx MCP Server": {
      "command": "/Users/yourname/jadx-mcp-plugin/venv/bin/python",
      "args": ["/Users/yourname/jadx-mcp-plugin/fastmcp_adapter.py"]
    }
  }
}
注意:您的密鑰屬於敏感信息,請勿與任何人分享。

替代品

M
MCP
微軟官方MCP服務器,為AI助手提供最新微軟技術文檔的搜索和獲取功能
8.7K
5分
A
Aderyn
Aderyn是一個開源的Solidity智能合約靜態分析工具,由Rust編寫,幫助開發者和安全研究人員發現Solidity代碼中的漏洞。它支持Foundry和Hardhat項目,可生成多種格式報告,並提供VSCode擴展。
Rust
5.9K
5分
D
Devtools Debugger MCP
Node.js調試器MCP服務器,提供基於Chrome DevTools協議的完整調試功能,包括斷點設置、單步執行、變量檢查和表達式評估等
TypeScript
5.4K
4分
S
Scrapling
Scrapling是一個自適應網頁抓取庫,能自動學習網站變化並重新定位元素,支持多種抓取方式和AI集成,提供高性能解析和開發者友好體驗。
Python
7.8K
5分
M
Mcpjungle
MCPJungle是一個自託管的MCP網關,用於集中管理和代理多個MCP服務器,為AI代理提供統一的工具訪問接口。
Go
0
4.5分
C
Cipher
Cipher是一個專為編程AI代理設計的開源記憶層框架,通過MCP協議與各種IDE和AI編碼助手集成,提供自動記憶生成、團隊記憶共享和雙系統記憶管理等核心功能。
TypeScript
0
5分
N
Nexus
Nexus是一個AI工具聚合網關,支持連接多個MCP服務器和LLM提供商,通過統一端點提供工具搜索、執行和模型路由功能,支持安全認證和速率限制。
Rust
0
4分
S
Shadcn Ui MCP Server
一個為AI工作流提供shadcn/ui組件集成的MCP服務器,支持React、Svelte和Vue框架,包含組件源碼、示例和元數據訪問功能。
TypeScript
12.1K
5分
B
Baidu Map
已認證
百度地圖MCP Server是國內首個兼容MCP協議的地圖服務,提供地理編碼、路線規劃等10個標準化API接口,支持Python和Typescript快速接入,賦能智能體實現地圖相關功能。
Python
26.2K
4.5分
M
Markdownify MCP
Markdownify是一個多功能文件轉換服務,支持將PDF、圖片、音頻等多種格式及網頁內容轉換為Markdown格式。
TypeScript
19.5K
5分
F
Firecrawl MCP Server
Firecrawl MCP Server是一個集成Firecrawl網頁抓取能力的模型上下文協議服務器,提供豐富的網頁抓取、搜索和內容提取功能。
TypeScript
64.8K
5分
S
Sequential Thinking MCP Server
一個基於MCP協議的結構化思維服務器,通過定義思考階段幫助分解複雜問題並生成總結
Python
21.8K
4.5分
M
Magic MCP
Magic Component Platform (MCP) 是一個AI驅動的UI組件生成工具,通過自然語言描述幫助開發者快速創建現代化UI組件,支持多種IDE集成。
JavaScript
16.5K
5分
N
Notion Api MCP
已認證
一個基於Python的MCP服務器,通過Notion API提供高級待辦事項管理和內容組織功能,實現AI模型與Notion的無縫集成。
Python
13.4K
4.5分
C
Context7
Context7 MCP是一個為AI編程助手提供即時、版本特定文檔和代碼示例的服務,通過Model Context Protocol直接集成到提示中,解決LLM使用過時信息的問題。
TypeScript
45.0K
4.7分
E
Edgeone Pages MCP Server
EdgeOne Pages MCP是一個通過MCP協議快速部署HTML內容到EdgeOne Pages並獲取公開URL的服務
TypeScript
16.4K
4.8分
AIBase
智啟未來,您的人工智慧解決方案智庫
© 2025AIBase