toutiao-search

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

• Third-party content exposure detected (medium risk: 0.30). The skill queries the Toutiao search API for third-party published articles and videos based on user-provided keywords, ingesting outsider-authored content titles and summaries into the execution context.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill contains instructions and code (references/alipay-recharge.md and SKILL.md) that instruct the agent to run npx -y @alipay/agent-payment@latest install from an external npm package source at runtime, which constitutes an unverifiable runtime external dependency from a weak/unknown provenance tier.

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

• Direct money access detected (high risk: 0.70). The skill documents a direct financial operation involving automated Alipay balance recharges ("当接口返回积分不足（code=3210）时，自动唤起支付宝完成充值后继续查询", cashierUrl, submit-payment, poll-payment-status). There is no established independent access control or service-enforced authentication gating the payment execution itself described in the documentation. Therefore, it is a direct operation without qualifying control, resulting in a High severity assessment.