Choose theme
AI Resources
REA
REA connects an AI coding agent to tools for investigating software: native binaries, JavaScript and Electron apps, .NET assemblies, and websites. You can also use its analysis tools from a terminal.
Its documented workflows connect findings to evidence and keep unresolved questions visible. Native analysis uses separate tools such as Hopper or Ghidra; it does not recover the original source code. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
App investigation tools for coding agents
A local command-line tool and MCP server for examining code, application packages, and selected runtime observations.
Why it stands out
Findings with a traceable basis
Results identify the artifact, analysis provider, relevant locations, and limitations, so a reader can distinguish a recovered clue from a remaining inference.
Availability
Public source and npm package
The repository declares MIT and provides the rea-agents package, setup documentation, and versioned releases. Its main supported hosts are macOS and Linux.
Why it matters
What makes it useful
When an application's source is unavailable, a developer can investigate how a feature is assembled. REA lets an agent search strings and functions, follow references and calls, and inspect decompiled routines before explaining the feature. The agent can then use its normal coding tools to build an implementation for the developer's own project.
What to know
Where it fits
REA supplies the inspection layer underneath a coding agent. Setup can configure Claude Code, Claude Desktop, Codex, Cursor, Gemini CLI, and Windsurf. Native decompilation needs Hopper or an existing Ghidra installation; individual operations have their own host and tool requirements.
Notable points
What stands out
Static JavaScript and Electron analysis maps modules, imports, routes, and IPC clues without running extracted modules. Runtime observation is a separate step. An apparent connection in recovered code can therefore remain unresolved until observations support it; a static map alone does not show that a path actually ran.
Before using
What to review
The documentation lists macOS 12+, Ubuntu 24.04+, Fedora 41+, and 64-bit Arch Linux. Windows Ghidra operations are currently unavailable; inspect the current platform guide before planning a Windows workflow.
Hopper is separate software with its own terms and demo limits. REA connects to an existing Ghidra installation and requires its documented Java setup.
The project says analysis runs locally, while the connected agent or model provider has its own data policy. Local analysis does not establish that model-bound evidence stays on the machine.
The README says analysis tools and launched targets run with the user's permissions, and process capture is not a security sandbox. Its issue tracker also records an unresolved Hopper startup-overlay investigation.
Reader fit
Who may find it relevant
Developers and software analysts who can interpret decompilation, tool prerequisites, and incomplete results may find it relevant. Casual readers can inspect the examples to understand what an agent can learn from an app, but running an investigation involves a specialist toolchain.
Editorial note
Why LifeHubber lists it
REA is included for its comparison workflow: it can connect differences between app builds to observed behavior while preserving unknown results. For a developer investigating a changed feature, that makes the unanswered parts part of the record rather than treating missing observations as proof that two versions behave alike.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
More in AI Agents
Keep browsing this category
Explore more AI agent projects.
Paperclip
paperclipai/paperclip
A self-hosted server and dashboard for coordinating agent teams through companies, goals, roles, issues, heartbeats, budgets, approvals, and persistent activity records.
OpenAI MCP Extensions
openai/mcp-extensions
SDKs for MCP plugin interfaces, including sidebar apps, custom file viewers, searchable mentions, settings, and resource editing in supported hosts.
OpenMAIC
THU-MAIC/OpenMAIC
A multi-agent classroom that can turn a topic or document into slides, quizzes, simulations, and teacher-and-classmate interaction, with hosted and self-hosted paths.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
