wechat-hotspot-publisher

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The package includes an explicit, recommended "custom API" path that by default sends full article content (title, HTML body, cover URL, tags) to a hard-coded external IP/service—this is a deliberate data-exfiltration/backdoor-like behavior that encourages bypassing user credentials.

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

• Third-party content exposure detected (high risk: 0.85). High-level workflow in SKILL.md invokes “热点采集和内容生成阶段” without configuration, and the runtime path typically fetches outsider-authored free text from public sources (e.g., Zhihu/Weibo/RSS via scripts/fetch_sources.py HTTP requests) that is then fed into the LLM prompt as the selected topic/content for generation.