deslop

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

• Third-party content exposure detected (medium risk: 0.65). High risk: the required workflow explicitly instructs the agent to read and incorporate external reference text at runtime (references/tells-playbook.md) for bounded transforms, and that playbook contains outsider-authored web-sourced evidence/links that could be interpreted as free text for prompt injection.