runtype-build-product

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

• Third-party content exposure detected (high risk: 0.85). The required workflow calls get_platform_documentation(...) and reads runtype://... MCP resources at runtime; these are platform docs provided by the Runtype MCP server (not authored by the operating user), so the LLM ingests outsider-authored free text via the MCP tool responses.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill declares a runtime MCP tool at https://api.runtype.com/v1/mcp/protocol which the agent calls (e.g., get_build_instructions, get_platform_documentation) to fetch live documentation/instructions that directly drive agent prompts and behavior.