OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Coding Agents

Reagent

Reagent is an open-source Python command-line agent that uses Ghidra and language models to reconstruct and validate C/C++ functions from compiled binaries.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Reagent — OpenViST project profile in Coding Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you have a compiled program to analyze, can provide Ghidra evidence and a supported language-model setup, and want candidate source recovery with explicit validation stages.

Who it is for

  • Reverse engineers examining compiled programs
  • C/C++ developers recovering or checking implementations against binary evidence
  • Analysts able to configure Ghidra, a supported model provider and project-specific validation

Documented uses

  • Gather binary-analysis evidence for selected functions.
  • Generate candidate C/C++ implementations with a language model.
  • Check candidates through model review, structural verification, configured build or test gates, and parity analysis.
  • Plan function work and track evidence or validation status through manifests.
Before you choose

Requirements & limits

The project is open source under the MIT license. The documented setup uses a local Python and Ghidra environment with Ghidra Bridge and a supported API or authenticated CLI model provider. Provider charges may apply.

  • Requires a target program and a configured Ghidra/Ghidra Bridge workflow.
  • Requires at least one supported model-provider setup; model usage may carry provider costs.
  • Project-specific source paths and validation commands need configuration.
  • A successful result depends on several distinct checks, and parity analysis is not a semantic-equivalence proof.
  • Generated candidate code is not automatically applied to the original source tree.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare Ghidra evidence and a project configuration

Set up the bridge evidence and Reagent configuration for a target project before starting reversal.

Before you start

  • Python 3.10 or newer
  • Git for installation from source
  • A local Ghidra installation and configured Ghidra Bridge
  • A supported provider: Claude or OpenAI-compatible API credentials, or an authenticated Claude or Codex CLI
  • A target program or project to analyze

Steps

  1. Install Reagent with the Ghidra Bridge extra using the documented PyPI command.

  2. From the target project, run ghidra-bridge init and edit the generated configuration with the Ghidra project and program paths.

  3. Run ghidra-bridge export all; the headless route requires the bridge headless extra and a local Ghidra installation.

  4. Run re-agent init --profile generic-cpp, then edit re-agent.yaml to select a model, set the bridge executable and source paths, and configure validation.

Expected outcome

The target project has bridge evidence exports and a Reagent configuration ready for the next reversal stage.

Guide limitations

  • The guide covers setup, not executing or validating a reversal.
  • Provider access and any model-call costs depend on the selected API or CLI setup.
  • Candidate validation and parity checks are not proof of semantic equivalence.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
Dryxio/reagent · Official homepage title · retained in its source language
How we review sourcesSuggest a correction