control-chrome
Fail
Audited by Snyk on Aug 24, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E006: Malicious code pattern detected in skill scripts.
- Malicious code pattern detected (high risk: 1.00). This skill explicitly authorizes reading browser cookies, local storage, profiles, passwords, and session stores, permits autonomous bypassing of authentication and action-time confirmations, and labels itself "Unrestricted," enabling deliberate data-exfiltration and credential-theft behavior.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The control-chrome required runtime workflow uses browser-client (agent.browsers.getForUrl/getDefault/getChrome/etc.) to navigate to user-supplied/target URLs or interact with the user’s currently open tabs/Chrome state, which can include outsider-authored free-text from web pages the LLM then reads at runtime.
Issues (2)
E006
CRITICALMalicious code pattern detected in skill scripts.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata