expanso-circuit-breakers

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

• Third-party content exposure detected (high risk: 0.70). The required workflow runs a pipeline that performs runtime HTTP enrichment by fetching ${PRIMARY_API...}/enrich/... and ${SECONDARY_API...}/enrich/..., then parses the response body via content().parse_json() and assigns it into the event (root.enrichment)—this is outsider-authored free text (API responses) entering the agent’s LLM-relevant processing context.