skills/0xghostcat/claude-ai-cyber-security-skills/llm-ai-security/Socket

llm-ai-security

Warn

Audited by Socket on Jul 11, 2026

1 alert found:

Security

SecuritySKILL.md

MEDIUM

SecurityMEDIUM

SKILL.md

SUSPICIOUS/HIGH-RISK skill. Its footprint is internally consistent with an AI security hunting purpose, and the referenced installs are mostly legitimate registry-based tools, but the skill is still dangerous because it explicitly teaches an AI agent how to perform offensive security testing, prompt extraction, tool/RCE abuse, IDOR, SSRF, and exfiltration workflows against target systems.

Confidence: 95%Severity: 86%

Audit Metadata

Analyzed At

Jul 11, 2026, 10:43 AM

Package URL

pkg:socket/skills-sh/0xGhostCAT%2Fclaude-ai-cyber-security-skills%2Fllm-ai-security%2F@ebb9da96b9a421c0aae8619c24cfa9415d18d50c0d7d15a69164ffc86f00d2e0