radare2

Warn

Audited by Socket on Aug 19, 2026

2 alerts found:

SecurityAnomaly
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill is internally coherent for radare2-based reverse engineering, and its referenced install source appears to be the official radareorg release channel. However, it grants an AI agent high-risk offensive analysis and binary-patching capability, includes auto-bootstrap of external tooling, and encourages immediate execution, so overall security risk is high even without clear evidence of credential theft or malicious exfiltration.

Confidence: 89%Severity: 81%
AnomalyLOW
scripts/recon.ps1

This script is primarily an orchestrator for running rabin2/r2 against a user-supplied file path. It does not show overt malware behaviors (no exfiltration, backdoors, or destructive actions) within the provided snippet. The dominant supply-chain risk is its automatic bootstrap mechanism for missing tools, which executes bootstrap-reverse.ps1 using powershell.exe with -ExecutionPolicy Bypass, combined with dot-sourcing and indirect tool command resolution from external helper code not shown here. Treat ToolDiscovery.ps1 and bootstrap-reverse.ps1 as high-trust dependencies and verify integrity and download/install behavior before use.

Confidence: 55%Severity: 65%
Audit Metadata
Analyzed At
Aug 19, 2026, 09:14 AM
Package URL
pkg:socket/skills-sh/zhaoxuya520%2Freverse-skill%2Fradare2%2F@cb7d317965d7aca92196c4e9e63f96085ffa5781cb67520fb1264a43fef9b667
Security Audit — socket — radare2