接送站助手

企业用车服务助手，支持即时用车、预约用车、接送机、包车等多种用车场景，提供车型选择、费用预估、订单管理等功能。Invoke when user needs to book a car, schedule a ride, airport transfer, or manage car service orders.

What to consider before installing: - The SKILL.md claims the skill MUST call real car‑service platform APIs and forbids fabricating prices or vehicles, but the included code simulates drivers and prices locally and contains no network calls, API endpoints, or credential handling. Ask the author to explain how real bookings are performed and to provide the actual integration code and required auth details. - The SKILL.md triggered a 'unicode control characters' scan hit; these characters can be used to hide or manipulate prompts. Inspect the SKILL.md file in a text editor that shows hidden characters and remove any suspicious control codes before use. - Running the included Python modules will execute local code that fabricates orders and uses random values; do not rely on this for actual bookings or billing. If you need production booking, require the skill to accept explicit API endpoints and API keys and show secure handling of those credentials. - Because there is no external install, the immediate risk is limited to whatever the Python code does locally. Still, review the full code for any unintended behaviors (networking, logging of sensitive data) before running in any environment that has production data. - If you are not the maintainer, treat this skill as a simulator/sample until the integration inconsistencies and hidden-character issue are resolved. If you want, request the author to: (1) remove hidden control characters, (2) provide documented external API endpoints and required env vars, and (3) include secure credential handling and network-call code if real bookings are intended.

Type: OpenClaw Skill Name: station-transfer Version: 1.0.0 The skill bundle provides a simulated car service assistant for booking rides and airport transfers. The Python implementation in `scripts/car_service_api.py` and `scripts/openai_adapter.py` uses mock data and randomized logic to simulate API responses without performing any actual network requests, file system operations, or sensitive data access. No evidence of malicious intent, data exfiltration, or prompt injection was found.