harness-pipeline

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text when users supply --jira <issue-key> and the skill then interacts with Jira issue details/comments (and posts comments, which can include user-provided content), making it possible for attacker-supplied text to be consumed indirectly during Jira validation/status/linking.