What problem it solves

Developers need to understand how an application feature works without access to its source code and reproduce that behavior in another project. REA helps investigate application behavior across binaries, application code, and runtime behavior, with results that include evidence and limitations behind each conclusion.

Key features

• Agent and terminal workflows — REA connects an agent through MCP and exposes the same workflows through a terminal CLI. It can register with supported agents and support direct inspection from the command line.

  Source: README, scripts/rea.mjs

• Broad target coverage — The project supports inspection of native binaries, JavaScript and Electron applications, .NET assemblies, websites, Android packages, firmware, captured network traffic, and numerous binary and artifact formats.

  Source: README, docs/native-investigation.md

• Evidence-backed findings — Findings include supporting evidence and stated limitations, allowing an agent to trace relevant code and use the results for follow-up questions, explanations, implementations, and tests.

  Source: README

• Local static and runtime analysis — Static JavaScript and .NET inspection reads supplied files without executing the application, while runtime capture can run or interact with a selected target under the user's permissions.

  Source: README

• Agent environment construction — Recorded API requests and responses can be converted into tools for an AI agent. The resulting environment uses recorded data and permits changes only when the recording confirms their effects.

  Source: README

Quick start

Copied verbatim from the project's README.

npx rea-agents setup

Understand how search works in the Notes app, show me the evidence, and build a
similar feature for my project.

npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json

Who it may suit

Inference: This project may suit developers, security researchers, and technical teams who need evidence-backed investigation of applications or binaries through an AI agent or terminal workflow.

Topics to discuss

• Which application or binary analysis workflow would be most useful to improve next?

• How should evidence, limitations, and unknowns be presented for complex investigations?

• What examples would best demonstrate turning an investigation into a tested implementation?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/morluto/rea#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.