google-cloud-solution-agentic-analytics-spark-knowledge-catalog

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md requires grounding generated content using the “Google Developer Knowledge MCP server” (developerknowledge:search_documents/get_documents/answer_query), which at runtime can fetch and inject third-party/free-form document text into the agent’s LLM context via the MCP tool outputs.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly requires grounding generated content by fetching documents from external services (e.g., the MCP server at https://developerknowledge.googleapis.com/mcp), which will be fetched at runtime and directly used to control the agent's prompts/outputs.