skill-kit-custom

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes custom Now Assist skill prompts that ingest outsider-authored free text at runtime via user_input and record_field inputs (e.g., incident_description/work_notes/category) that are injected into LLM templates like “Description: {incident_description}” and “Work Notes: {work_notes}”.