review-korean-patent-claims-ip

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow ingests and extracts “complete supplied files” including submitted claims/specification/office actions and any retrieved patent evidence for analysis (Step 1 and Step 3/5), meaning an outsider can provide free text via the required inputs/files that the agent reads directly.