leanrea vs rea
Reverse-engineering guidance and tools for AI agents
leanrea and rea help coding agents investigate software artifacts and trace application behavior. leanrea provides compact guidance for agents to run command-line tools, while rea connects agents to analysis workflows through an MCP server and CLI.

leanrea: Reverse Engineer Software with Agent Skills and CLI Tools
leanrea is an 8 KB skill that guides coding agents through software reverse engineering with ordinary command-line tools instead of an MCP server. It provides target-specific commands and workflows for inspecting apps, binaries, bundles, and other artifacts.

rea: Reverse Engineer Apps and Binaries with AI Agents
REA is a TypeScript MCP server and CLI that helps AI agents investigate application behavior, native binaries, and runtime activity. It returns evidence-backed findings for targets including JavaScript and Electron apps, Android packages, websites, and native code.
| leanrea | rea | |
|---|---|---|
| Language | Python | TypeScript |
| License | MIT | MIT |
| Stars | 372 | 428k |
| Forks | 75 | 107k |
| Last analyzed | Oct 17, 2026 | Nov 4, 2026 |
Key differences
- leanrea is an 8 KB skill that directs agents to ordinary CLI tools; rea is a TypeScript MCP server and CLI that connects agents to analysis providers.
- leanrea emphasizes suggested commands and workflows for artifact types; rea also supports runtime and device investigations across application layers.
- leanrea is written in Python, while rea is written in TypeScript. Both are MIT licensed.
- leanrea's reported benchmark compares it with REA on a small set of tasks; rea's broader tool support can require caller-supplied providers and varies by target.
- leanrea is described as very new, with work concentrated in one contributor; rea is described as actively maintained, with frequent releases and a broader contributor base.
Choose leanrea if you…
- want compact investigation guidance without adding an MCP server.
- prefer to have an agent run locally available CLI tools through Bash.
- need first-pass workflows for artifacts such as Electron apps, binaries, firmware, or databases.
Choose rea if you…
- need an MCP server and CLI for agent-driven analysis workflows.
- want to investigate runtime behavior, devices, or interactions across application layers.
- need evidence-backed findings from supported analysis providers, such as pseudocode, assembly, or code references.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.