rhdh-jira-refine

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required workflow ingests outsider-authored free text from the caller by accepting either issue keys/JQL/sprint/backlog (which determines which Jira issues and their fields/comments are fetched and analyzed), so an outsider can supply a crafted JQL/selection that causes the agent to read attacker-controlled Jira issue/comment content at runtime.