sci-download

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt shows examples that inline API keys and cookies in commands and config updates (e.g., python -c "…update_config('elsevier_api_key','YOUR_KEY')") and instructs exporting cookies to files, which requires the agent to accept and embed secret values verbatim — an insecure pattern.

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

• Third-party content exposure detected (high risk: 0.75). At runtime the skill fetches outsider-authored free text from public sources (e.g., arXiv Atom feed via requests.get(https://export.arxiv.org/api/query) and CNKI HTML via requests.post(...)), then parses and includes fields like title/abstract/error messages into the returned JSON that is printed/consumed by the agent.