deity-agent-builder

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly includes WebSearch and WebFetch built-in tools (see reference/tools-reference.md, SKILL.md Step 4 and the templates that register / ) which fetch public URLs and whose outputs are consumed by agents (tools registered in the LLM loop, Result/Observe/Validate logic and workflow items), so untrusted third‑party page content can be read and materially influence LLM decisions and subsequent tool use.