empathy-analysis

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly states "Accept transcripts, interviews, pasted content, files, and URLs as input — ingest them, don't summarize them," which requires the agent to read and interpret arbitrary user-provided or third‑party URLs/transcripts and use that content to build empathy maps and guide decisions, exposing it to untrusted user-generated content that could carry indirect prompt injection.