re-android-crypto
Warn
Audited by Socket on Sep 14, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally coherent as an Android crypto-audit/reverse-engineering workflow and uses mostly legitimate local tools, but it equips an AI agent with high-risk offensive instrumentation capability against Android apps. The main issue is that this is a security/exploit-style skill, plus a mild mismatch between declared `key-extraction` capability and the documented boundary against extracting key bytes. No clear credential harvesting, remote exfiltration, or malicious installer behavior is present.
Confidence: 88%Severity: 76%
Audit Metadata