qa-response

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

• Third-party content exposure detected (high risk: 0.70). The skill’s runtime input is $ARGUMENTS (the drafted customer response text), which is not shown to be authored by the operating user; if $ARGUMENTS contains customer-authored free text, it will be ingested into the LLM context for review (outsider free-form message/body text path).