browse-stealth

Warn

Audited by Socket on Apr 14, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill is internally consistent with its stated purpose, but that stated purpose is itself high-risk: enabling an AI agent to stealth-browse arbitrary sites and bypass anti-bot defenses using persistent sessions and proxies. There is no clear credential-harvesting endpoint or overt malware behavior, but the combination of anti-detection guidance, arbitrary web access, authenticated session handling, and Bash execution makes this a high-risk browsing capability that should not be broadly trusted.

Confidence: 89%Severity: 84%
Audit Metadata
Analyzed At
Apr 14, 2026, 06:31 AM
Package URL
pkg:socket/skills-sh/ulpi-io%2Fskills%2Fbrowse-stealth%2F@61dd61fa3e36de14519212806296e4024574c87d