wechat-clippings-vp

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt includes an explicit, insecure example that instructs passing the TikHub API key as a literal CLI argument (--tikhub-api-key "<tikhub_api_key>"), which would require the LLM to emit secret values verbatim if used that way.

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/clip_wechat_tikhub.py, the runtime ingests outsider-controlled free text via --input/--title-source/query fields to discover WeChat articles and then fetches and converts the resulting article HTML (tikhub_get → fetch_article_detail → html_to_markdown → BodyHTMLParser), so poison text can reach the LLM through TikHub-returned #js_content/content.