OpenViSTAI · Agents · Blockchain · Crypto
Menu
MCP & Tooling

Jadx AI MCP

Jadx AI MCP combines a plugin for the JADX Android decompiler with an MCP server, allowing MCP-compatible LLM clients to interact with APK analysis workflows. It is designed to assist with inspecting APKs, reviewing decompiled code, vulnerability analysis and reverse engineering.

Independent OpenViST project profile, based on linked sources.

Jadx AI MCP — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you already work with JADX and want an MCP-enabled LLM client to assist with APK inspection or code review.

Who it is for

  • Android security researchers and reverse engineers using JADX.
  • Developers who want to connect an MCP-capable LLM client to JADX analysis.

Documented uses

  • Connect an LLM client to JADX through MCP.
  • Use conversational prompts to help inspect APKs and decompiled code.
  • Support vulnerability analysis and Android application reverse engineering.
Before you choose

Requirements & limits

  • The integration requires JADX, the plugin and MCP server, plus an MCP-compatible client.
  • The README lists Java 11+ and Python 3.10+; its server setup uses uv.
  • The README documents Claude Desktop, Cherry Studio and LM Studio configuration, along with stdio and HTTP modes.
  • HTTP binding to non-localhost interfaces is unauthenticated and unencrypted; the project advises restricting network access or using an SSH tunnel.
  • The repository lists an Apache-2.0 license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up JADX-AI-MCP with Claude Desktop

Install the JADX plugin and MCP server files, then configure Claude Desktop to start the server.

Before you start

  • Java 11 or later and Python 3.10 or later, as indicated by the project badges.
  • JADX and Claude Desktop with MCP enabled.
  • The uv tool for server dependency management.
  • The JADX-AI-MCP release JAR and server ZIP, unless installing the plugin through the documented JADX CLI method.

Steps

  1. Download the jadx-ai-mcp-.jar and jadx-mcp-server-.zip from the project releases, then unpack the server ZIP. Alternatively, install the latest plugin with jadx plugins --install "github:zinja-coder:jadx-ai-mcp".

  2. For the unpacked server, enter its jadx-mcp-server directory. Install uv if needed; if dependency errors occur, the README lists the command below, optional environment activation, and optional installation of httpx and fastmcp.

    Documented command
    uv venv
  3. Open Claude Desktop’s MCP configuration file: ~/.config/Claude/claude_desktop_config.json on Linux, %APPDATA%\Claude\claude_desktop_config.json on Windows, or ~/Library/Application Support/Claude/claude_desktop_config.json on macOS. Add an mcpServers entry named jadx-mcp-server using uv as command and arg

  4. Replace the example uv and server-directory paths with their actual paths, then use Claude Desktop to interact with the integration.

Expected outcome

The README describes setup of the JADX plugin and MCP server in Claude Desktop for interaction with JADX through MCP.

Guide limitations

  • The README’s GUI installation alternative is shown through images; the text does not specify its individual clicks.
  • The README’s project badges specify Java 11+ and Python 3.10+; compatibility beyond those badges is not stated.
  • This setup guide does not establish that a particular APK analysis or vulnerability finding will succeed.
Evidence and transparency

Sources & review

Source record details
Source availability
Active
Sources checked
9 October 2026
Original website title
zinja-coder/jadx-ai-mcp · Official homepage title · retained in its source language