wc-signal-emitter

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

• Third-party content exposure detected (high risk: 0.70). The skill’s runtime workflow ingests a “signal (frontmatter + body)” provided by the caller/other agent, and that body is free text that can be authored by an outsider; if it passes validation it is persisted for downstream LLM reading, creating indirect prompt-injection risk via outsider-authored prose.