ghqr-report

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md Phase 1 reads “ghqr JSON output” from either a user-provided scan result file/JSON path or the ghqr_scan tool output file URI, and Phase 2 then renders fields like evaluation.recommendations[].issue and ...recommendation directly into the report, so outsider-authored text can enter via the scan result content.