{"slug": "rea-ask-your-coding-agent-how-a-finished-app-works", "title": "REA: ask your coding agent how a finished app works", "summary": "An open-source project called REA (Reverse Engineer Anything) provides an MIT-licensed MCP server that connects coding agents such as Claude Code, Codex, Cursor and Gemini CLI to reverse-engineering tools for inspecting finished applications. The server traces the code behind a requested feature across native binaries, JavaScript and Electron apps, .NET assemblies, Android APKs and websites, returning each finding with supporting evidence and stated limitations. A showcase reconstruction of a DX-Ball sound-pan function in C passed 3,205 test cases against the original x86 code and reproduced all 63 compiled bytes of the function.", "body_md": "I recently came across [REA](https://github.com/morluto/rea), short for Reverse Engineer Anything. It connects a coding agent to tools for inspecting a finished app, so you can ask how a particular feature works.\n\nREA is an open-source MCP server under the MIT license. It connects Claude Code, Codex, Cursor, Gemini CLI and other agents to reverse-engineering tools. You point the agent at an app, ask about one feature, and it traces the code behind it.\n\n*I haven't run it myself yet. Everything below comes from the project's documentation, plus how I plan to check it.*\n\nThe documented targets include:\n\n| Target | What comes back | \n|---|---|\n| Native binaries | Pseudocode, assembly, strings, symbols, calls and references (Hopper, Ghidra or IDA for deep analysis) | \n| JavaScript and Electron apps | Modules, imports, source maps, routes and IPC | \n| .NET assemblies | Metadata and CIL instructions, static analysis only | \n| Android APKs | Manifest, classes and decompiled methods (needs JADX and a JDK) | \n| Websites | Page structure, scripts and network observations | \n\nWith Node.js and npm installed, run:\n\n```\nnpx rea-agents setup\n```\n\nReview and approve the proposed configuration changes, then restart your agent. Setup registers the MCP server and installs matching workflow instructions. Static JavaScript and .NET analysis work without a native engine. The FAQ says analysis runs locally, although your agent still sends the tool results to its model provider.\n\nREA returns each finding together with the evidence and the limitations behind it. [One showcase](https://rea.tools/showcase/notion/) follows Notion's clipboard API from the renderer, through preload and IPC, into the main process. \n\nIn [another](https://rea.tools/showcase/dx-ball/), REA supplies the instructions and data reads behind a sound-pan function in the old game DX-Ball, and a reconstruction project rewrites it in C. That version passes **3,205 test cases** against the original x86 code and reproduces **all 63 compiled bytes** of the function.\n\nThat's the level of detail I want from an engineering agent. You can follow the path behind a feature and ask where each conclusion came from.\n\nThe example prompt in the README asks the agent to understand how search works in the Notes app, show the evidence, and build a similar feature for your project. That gives the agent a way to use what it finds when writing an implementation. I'd still test that implementation against the behavior I wanted to reproduce.\n\nThe README is explicit that REA is for lawful research, interoperability, software preservation, and software you own or are authorized to examine. Commercial tools such as IDA, Hopper and JEB need your own valid licenses.\n\nInstall it only from the official sources (the [GitHub repo](https://github.com/morluto/rea), [rea.tools](https://rea.tools/) or the `rea-agents` npm package), because the project warns about look-alike repositories.\n\nI'd start with one narrow question on an Electron app I'm authorized to inspect: which code path produces this behavior? Then I'd check whether another engineer can reproduce the finding from REA's evidence alone, and whether the app behaves the way the explanation predicts when I change the input.", "url": "https://wpnews.pro/news/rea-ask-your-coding-agent-how-a-finished-app-works", "canonical_source": "https://dev.to/apovalov/rea-ask-your-coding-agent-how-a-finished-app-works-1ebd", "published_at": "2026-10-17 09:51:16+00:00", "updated_at": "2026-10-17 10:03:06.322899+00:00", "lang": "en", "topics": ["ai-agents", "agent-protocols", "developer-tools", "ai-tools"], "entities": ["REA", "Claude Code", "Codex", "Cursor", "Gemini CLI", "Notion", "DX-Ball", "Ghidra"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/rea-ask-your-coding-agent-how-a-finished-app-works", "markdown": "https://wpnews.pro/news/rea-ask-your-coding-agent-how-a-finished-app-works.md", "text": "https://wpnews.pro/news/rea-ask-your-coding-agent-how-a-finished-app-works.txt", "jsonld": "https://wpnews.pro/news/rea-ask-your-coding-agent-how-a-finished-app-works.jsonld"}}