data-engineering-best-practices

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflows (see SKILL.md Inputs to Collect and the PR_REVIEW/AIRFLOW mode descriptions) explicitly ask for "PR diff or link" and "DAG code or file path to review", meaning the agent will ingest and act on arbitrary user- or public-hosted PRs/code (e.g., GitHub links), which are untrusted third-party content that can materially influence decisions and tool use.