re-agent: evidence-backed reverse-engineering skills
A modular set of coding-agent skills for evidence-backed static, runtime and memory analysis, for authorised and legal targets.
Credits
Earliest dated source:
The full story
RE-Agent is a fork of REA refocused on maintaining authorised legacy software: desktop apps, command-line tools and games that are no longer supported. It's a local-first, evidence-driven toolkit that connects coding agents to static analysis, controlled runtime observation, memory evidence, and limited reconstruction and verification workflows through CLI and MCP interfaces. It works with tools such as Ghidra, IDA, Binary Ninja, Frida and Cheat Engine.
It's in active development, and is clear that it doesn't recreate arbitrary applications automatically or prove that a whole program is equivalent. What can be verified depends on scope, platform and the observations available.
How it was done
Tools
Themes
Loading themes…
Sources
From the maker's own write-up or code. We summarise what they report; we haven't tested it ourselves.
- REA community: heaven-hm/re-agent: Reverse Engineering Agent: modular coding-agent skills and evidence-backed statiRepository account: heaven-hmPublished
The fork's README, with its workstreams and the evidence each one needs.