adk

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests and uses public website content (e.g., "adk kb sync" and "Website" sources in references/cli.md and the use of WebsiteKB in references/conversations.md and SKILL.md), so the agent may fetch arbitrary web pages and incorporate their untrusted content into RAG/execution flows, enabling indirect prompt injection.