didi-ride-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to ask the user for their MCP Key and to run a CLI command that embeds the provided key verbatim (openclaw config set '...','YOUR_KEY'), which requires the LLM to handle and output secret values directly.

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

• Third-party content exposure detected (medium risk: 0.30). 该技能在运行时会直接读取并处理“用户对话/消息文本”（触发词如“打车去[地点]”“查询订单/查询订单 ”“15分钟后打个车”等均来自外部用户输入），并将其中的自由文本（起终点、订单号等）用于调用 MCP 工具，从而存在间接提示注入风险。