linkfox-onboarding

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to call scripts that return an "api_key" and then "拿到 api_key 后... 帮用户配置" environment-variable-setting commands (e.g., setx / echo export), which requires the LLM to read and embed the secret API key verbatim into its output—an explicit secret-exfiltration pattern.

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

• Direct money access detected (high risk: 1.00). This skill explicitly implements a recharge/payment flow: it lists purchasable plans, lets the user select a plan and payment method (wechat/alipay), calls a script to create an order (python scripts/create_order.py) which returns order_id, qr_content, qr_url, pay_url and PNG QR, and provides a script to query payment status (python scripts/query_order.py). Because it contains specific payment gateway interactions and order-creation APIs (not just generic HTTP or browsing), it grants direct financial execution capability.