optimizing-queries

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly reads and analyzes arbitrary user- or third-party-provided SQL, EXPLAIN outputs, migration and code files (see scripts like scripts/analyze_query.py, scripts/parse_explain.py, scripts/detect_orm_issues.py which accept file paths or stdin), so the agent will ingest and interpret untrusted content as part of its workflow.