dx-data-navigator

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

• Third-party content exposure detected (high risk: 0.80). The skill queries user-generated content tables such as github_issues, confluence_pages/confluence_page_versions, and incidents (source_url) from the DX database—i.e., ingested public GitHub/Confluence/incident content—which the agent is expected to read and interpret as part of its workflow, enabling indirect prompt injection risk.