data-dict-extractor

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

• Third-party content exposure detected (high risk: 0.75). The skill reads outsider-authored free text from the Drupal config export YAML files in the user-supplied --sync-dir (e.g., node.type.*.yml, field.field.*.yml, core.entity_form_display.*.yml) via load_yaml() and then injects it into the LLM context indirectly by returning it as generated Excel content (cell values like labels/descriptions/widgets), which are derived from that external directory’s contents.