subkeyword-injector
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and snapshot arbitrary public web pages using a Browser MCP or the agent-browser commands (e.g., "agent-browser open " and snapshot) to extract headings/content, which exposes it to untrusted third-party web content that could carry indirect prompt injection.
Audit Metadata