browser-mcp-agent
Warn
Audited by Socket on Aug 28, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's capabilities match its stated browser-automation purpose and its docs are unusually candid about risks, but it still hands an agent powerful logged-in browser control and relies on a closed-source vendor binary fetched at runtime. Because that unverifiable binary receives credentials/license traffic, the skill carries high security risk even without evidence of confirmed malware.
Confidence: 88%Severity: 84%
Audit Metadata