xiaohongshu-darkhorse

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

• Third-party content exposure detected (medium risk: 0.30). The workflow fetches Xiaohongshu posts from an external API (redfox.hk), where post titles and descriptions written by arbitrary users are ingested and processed by the agent.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill makes runtime requests to a third-party API endpoint on redfox.hk (https://redfox.hk/story/api/cozeSkill/getLowPowderExplosiveArticle) and loads external client-side scripts from unversioned/common CDNs (jsdelivr.net, cdnjs.cloudflare.com, unpkg.com), but lacks a clearly evidenced strong official first-party publisher provenance or enforced cryptographic verification.