skills/google/oss-fuzz/fuzzing-memory-unsafe-expert/Socket

fuzzing-memory-unsafe-expert

Warn

Audited by Socket on Jul 30, 2026

1 alert found:

Security

SecuritySKILL.md

MEDIUM

SecurityMEDIUM

SKILL.md

SUSPICIOUS: the skill is internally consistent and not credential-stealing, but it grants an AI agent offensive fuzzing capability and instructs local compilation/execution of arbitrary target code. Main risk is operational abuse and host exposure, not malware or exfiltration.

Confidence: 91%Severity: 72%

Audit Metadata

Analyzed At

Jul 30, 2026, 11:39 AM

Package URL

pkg:socket/skills-sh/google%2Foss-fuzz%2Ffuzzing-memory-unsafe-expert%2F@370af868a997562fe8e28930432f53f56a2414783ff2de520404335c8065a586