skills/bleedingdev/agentic-security-workspace/web-security-assessment/Socket

web-security-assessment

Warn

Audited by Socket on Oct 2, 2026

1 alert found:

Security

SecuritySKILL.md

MEDIUM

SecurityMEDIUM

SKILL.md

Purpose and capabilities are internally consistent for a pentest skill, and the referenced tools are plausibly official. Risk is still high because it equips an AI agent with offensive security capabilities and real-world scanning actions, with moderate supply-chain exposure from external tool installs and runtime updates.

Confidence: 89%Severity: 81%

Audit Metadata

Analyzed At

Oct 2, 2026, 12:19 PM

Package URL

pkg:socket/skills-sh/bleedingdev%2Fagentic-security-workspace%2Fweb-security-assessment%2F@70c6e215f1ab9b6c92cab860fbe91226a4e312a76f29bef034080f5dfe8c1ee7