wechat-write

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

• Third-party content exposure detected (medium risk: 0.30). The workflow queries and ingests trending articles from WeChat Official Accounts via the RedFox data API, which include article titles, summaries, and content authored by third-party public account operators.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill executes scripts/fetch_gzh_trends.py which makes runtime API requests to redfox.hk, an unverified commercial third-party platform of weak provenance.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 0.70). The skill documents a WeChat article generation workflow relying on an API that automatically triggers Alipay recharges (cashierUrl, amount, outTradeNo) when API credits are insufficient (exit code 3), orchestrating payment submission and polling via alipay-bot commands. This constitutes a direct financial operation (charges/payments) managed by the agent workflow without independent access control enforced by the service or execution layer at payment time.