Rewind
LLM-enhanced decompiler for Solana programs enabling reverse engineering and security analysis of deployed on-chain bytecode
Overview
Rewind leverages open and close source tools as well as large language models to decompile deployed Solana programs, transforming sBPF bytecode into various Intermediate Language representations for enhanced program analysis.
Why Rewind?
Security researchers and auditors frequently encounter deployed programs without access to source code, making vulnerability analysis and security verification fairly complex through traditional means.
Rewind solves this by employing LLM-enhanced decompilation and custom open-source and closed source tools to transform bytecode into readable Rust-like pseudo-code, enabling better understanding of closed-sourced programs deployed on-chain.
Rewind can enable security teams to gain program logic understanding quicker and verify implementation details without depending on external documentation or source code availability.
How It Works
Select On-chain Program
Download desired program to analize from the Solana blockchain
Analysis Pipeline
Our analysis pipeline analyzes strings and other artifacts and uses additional tools, populating context to enhance code understanding
Decompilation Pipeline
Our multi-layered decompilation pipeline generates pseudo-rust from different lifting layers and guided LLM augmentation
Annotate
Rename variables, apply types or regenerate analysis
Key Features
Reverse Engineering capabilities for Solana programs
sBPF Decompilation
Native support for Solana eBPF bytecode decompilation and analysis
LLM-Enhanced Analysis
Advanced language models for intelligent pattern recognition and code understanding
Custom Analysis Pipeline
Tailored analysis for Solana-specific artifacts and patterns
Use Cases
Essential tool for security researchers and developers working with deployed programs
Security Analysis
Incident Response
Due Diligence
Understand Deployed Smart Contracts
Request early-access to our decompilation tools