jadx-ai-mcp
Plugin for JADX to integrate MCP server
JADX-AI-MCP is an MCP server and JADX decompiler plugin that enables AI-powered analysis of Android APK files through the Model Context Protocol. It allows LLMs like Claude to interact with decompiled Android code in real-time, providing capabilities for security analysis, vulnerability detection, code review, and reverse engineering tasks. The system consists of two components: a JADX-GUI plugin that exposes decompiled APK data via HTTP, and a Python-based MCP server that translates between LLM clients and the JADX plugin. Developers can use it with Claude Desktop, Cherry Studio, LM Studio, and other MCP-compatible clients to perform automated static analysis, rename obfuscated code, analyze Android manifest files, and debug APKs with AI assistance.
Key Features
Use Cases
- 01Security researchers analyzing Android APKs for vulnerabilities, hardcoded secrets, and insecure API usage
- 02Reverse engineers deobfuscating and understanding obfuscated Android applications with AI assistance
- 03Mobile security auditors performing automated static application security testing (SAST) on APK files
- 04Developers analyzing third-party libraries and SDKs integrated into Android apps
- 05Malware analysts investigating suspicious Android applications for malicious behavior patterns
- 06Penetration testers identifying Android permission abuse and privacy leaks during security assessments
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
jadx-ai-mcp — FAQ
What is JADX-AI-MCP and what does it do?+
JADX-AI-MCP is an MCP server and JADX decompiler plugin that connects AI assistants like Claude to decompiled Android APK code. It enables real-time code analysis, vulnerability detection, and reverse engineering tasks through natural language prompts.
How do I install and configure JADX-AI-MCP?+
Install the JADX plugin using 'jadx plugins --install github:zinja-coder:jadx-ai-mcp', then set up the Python MCP server using uv. Configure your MCP client (Claude Desktop, LM Studio, etc.) to point to the jadx_mcp_server.py script with the appropriate command and arguments in the client's MCP configuration JSON.
Which AI clients work with JADX-AI-MCP?+
JADX-AI-MCP works with Claude Desktop, Cherry Studio, LM Studio, and any MCP-compatible client that supports stdio or HTTP transport modes.
Do I need API keys or paid services to use this?+
You need an API key for the LLM provider you choose (e.g., Anthropic Claude API). The JADX-AI-MCP server and plugin themselves are free and open-source under Apache 2.0 license. You'll also need Java 11+ for the JADX plugin and Python 3.10+ for the MCP server.
What are the prerequisites for running JADX-AI-MCP?+
You need JADX decompiler (GUI version), Java 11 or higher, Python 3.10 or higher, and the uv package manager. The system runs locally by default with the plugin listening on port 8650 and MCP server on port 8651.
Can I run JADX-AI-MCP remotely or in a container?+
Yes, the MCP server supports custom host/port configuration using --host, --port, --jadx-host, and --jadx-port flags. You can run the JADX plugin on one machine and the MCP server on another, or expose the MCP server from a Docker container using --host 0.0.0.0, though this requires proper network security considerations.
How do I install jadx-ai-mcp?+
Open the source repository on GitHub and follow its README. jadx-ai-mcp is a mcp server — MCP Agents Market links you directly to the official repo.
Is jadx-ai-mcp free?+
jadx-ai-mcp is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.