testing-websocket-api-security

Warn

Audited by Socket on Mar 15, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill’s purpose and capabilities are aligned, but that purpose is offensive security testing. It enables an AI agent to run exploit-style WebSocket assessments, includes explicit data-exfiltration PoC behavior, and performs DoS actions against external targets. Tool install sources are mostly normal, so the main risk is operational misuse and harmful network actions rather than supply-chain deception.

Confidence: 95%Severity: 93%
Audit Metadata
Analyzed At
Mar 15, 2026, 12:28 AM
Package URL
pkg:socket/skills-sh/mukul975%2FAnthropic-Cybersecurity-Skills%2Ftesting-websocket-api-security%2F@ecb27d2093b721f612234e149b241c945ab76ee5