hunt-nodejs

Warn

Audited by Socket on Sep 29, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

High-risk offensive security skill. It is internally consistent with its stated purpose, but that purpose is to direct an AI agent to perform exploitation and exfiltration against Node.js targets, including RCE validation and secret extraction via collaborator endpoints.

Confidence: 97%Severity: 94%
Audit Metadata
Analyzed At
Sep 29, 2026, 06:50 AM
Package URL
pkg:socket/skills-sh/murrtada%2Fbug-bounty-agent-skills%2Fhunt-nodejs%2F@625d5c88945776ee582fb9261846c31529682f3715062c6caa2df353e3595ef6