doc-processing

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path can fetch a user-supplied URL (public web content) via doc-download.py, then doc-convert.py converts the downloaded document into readable markdown (content.md) that is subsequently read to enrich meta.md, so outsider-authored free text can enter the agent’s LLM context through the document’s body text.