paper-context-resolver

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly directs the agent to consult external paper sources (the README-provided paper link, official project/paper pages, arXiv/OpenReview, and Google Scholar) to fetch and interpret public third-party content to resolve reproduction-critical details, so untrusted web content can materially influence agent decisions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and use external paper links at runtime (e.g., arXiv/OpenReview URLs such as https://arxiv.org/abs/xxxx.xxxxx) to resolve reproduction-critical details, which means externally fetched content would directly control the agent's prompts/outputs.