crw-research

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow calls fastcrw’s Research API to retrieve arXiv paper metadata (e.g., via results[].ids.arxiv from /v2/search/research/papers?... and citation-graph endpoints), which is outsider-authored free text/content originating from public arXiv pages that the LLM ingests as retrieved results.