security-audit

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow can fetch and ingest free-form text from an outsider-supplied URL (“passive”/“standard”/“active” modes include “one fetch of the supplied URL” and, in active mode, crawling/scanning of the target), which can include arbitrary third-party page content that the LLM may summarize into its context.