agent-sdk

Warn

Audited by Snyk on Feb 16, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). The skill explicitly exposes WebSearch/WebFetch capabilities (see "Built-in capabilities: Web search
  • Access real-time information" and the "Tool Mapping" entry "Web Search | WebSearch, WebFetch"), meaning the agent can fetch and read arbitrary public web content which is untrusted and can be interpreted at runtime, enabling indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 11:57 AM