wechat-similar-account

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

• Third-party content exposure detected (high risk: 0.85). Runtime path: the skill calls RedFox’s public API (POST https://redfox.hk/story/api/gzhUser/querySimilarAccounts), and the returned JSON includes outsider-authored free text fields like article title/summary inside works, which the script formats into LLM-visible output (e.g., “近5篇文章” and recommendation reasons).

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (2 type(s) found)