agentbay-monitor-skills

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

• Third-party content exposure detected (medium risk: 0.30). The workflow monitors and crawls social media platforms and web searches for keywords (via scripts/crawl.py and platform adapters), ingesting outsider-authored posts and articles into JSON data files that are subsequently processed and summarized by the agent.

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

• Potentially malicious external URL detected (medium risk: 0.30). The code utilizes urllib.request.urlopen to download a fallback font from cdn.jsdelivr.net during PDF export generation, which constitutes an unverifiable runtime external dependency under common/reputable unofficial provenance.