theorize-metaphysics

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the engine to "use web access when available" (SKILL.md) and to record and use "external sources" and "web search when available" in the repository workbench (references/repo-lab-protocol.md and source-map.md), which means the agent will fetch and interpret open/public third‑party content as part of its reasoning and decision-making, exposing it to untrusted user-generated inputs that could carry indirect prompt injections.