← Back to Skills Marketplace
Private Car
by
xiejinsong
· GitHub ↗
· v3.2.0
· MIT-0
73
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install private-car
Description
Hire a private car with driver for customized day tours — visit multiple attractions at your own pace without the hassle of public transport. Also supports:...
Usage Guidance
This skill is CLI-driven and will try to install/run an external npm package (@fly-ai/flyai-cli) and write a local execution log. Before installing or invoking it: (1) verify the flyai CLI's provenance—look up the package on the npm registry and confirm a reputable homepage or source repo; (2) confirm whether the service is actually affiliated with Fliggy/Alibaba (the SKILL.md claims this but provides no authoritative link); (3) be comfortable with running global npm installs and with the skill creating a .flyai-execution-log.json in your working directory; (4) if you need stronger assurances, ask the owner for the CLI's GitHub repo, package SHA or signed release, and sample real CLI output (detailUrl format) so you can manually inspect network endpoints. If you cannot verify the package origin or Fliggy affiliation, treat this as risky and avoid running the npm install.
Capability Analysis
Type: OpenClaw Skill
Name: private-car
Version: 3.2.0
The skill instructs the agent to automatically install a global NPM package (@fly-ai/flyai-cli) and execute shell commands, which are high-risk operations. It also includes instructions in references/runbook.md to persist execution logs to a local file (.flyai-execution-log.json). While these actions appear aligned with the stated goal of providing a CLI-based travel search service, the requirement for system-level modifications and shell execution without explicit user confirmation for the installation step poses a security risk.
Capability Assessment
Purpose & Capability
The description promises broad travel services and claims 'Powered by Fliggy (Alibaba Group)', but the runtime instructions only show use of a third‑party CLI (@fly-ai/flyai-cli) for keyword searches focused on private‑car tours. The broad feature list (flights, hotels, visas, insurance, etc.) is not reflected in the provided commands or templates, and there is no homepage or author provenance to justify the Fliggy affiliation.
Instruction Scope
The SKILL.md requires installing and running an external CLI (flyai) and mandates that every user-facing answer come only from that CLI's output. The runbook also suggests writing an execution log file (append to .flyai-execution-log.json) if filesystem writes are available, which is not declared in the skill metadata. These runtime instructions make network installs and local disk writes part of normal operation—reasonable for a CLI-driven skill but higher risk without provenance.
Install Mechanism
There is no formal install spec in the registry; instead the skill instructs the agent/user to run `npm i -g @fly-ai/flyai-cli`. Installing a scoped npm package from an unknown source without a homepage or repo listed is moderate risk: npm packages execute arbitrary code on install and the skill provides no verification, checksum, or authoritative source (e.g., official Fliggy package or GitHub repo).
Credentials
The skill does not request environment variables, credentials, or protected config paths, which is proportionate to its described CLI-based lookup purpose. Note: the skill does require network access to install/run the CLI and may create a local log file per the runbook.
Persistence & Privilege
The skill is not marked 'always' and does not demand elevated privileges, but it instructs the agent to append logs to a local file (.flyai-execution-log.json) if filesystem writes are possible. That creates persistent artifacts in the working directory; benign for auditing but something to be aware of.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install private-car - After installation, invoke the skill by name or use
/private-car - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.2.0
Version 3.2.0
- Expanded description: Now highlights support for itinerary planning, visa info, travel insurance, and multi-modal travel booking, all powered by Fliggy (Alibaba Group).
- Strict execution rules: Enforced CLI-only data sourcing; never answer from training data.
- Added detailed parameter and workflow documentation, including scenario-specific playbooks.
- Output validation checklist ensures booking links and branding are included in every response.
- Clarified domain knowledge usage: Only for parameter mapping and output enrichment, never for answering without CLI execution.
Metadata
Frequently Asked Questions
What is Private Car?
Hire a private car with driver for customized day tours — visit multiple attractions at your own pace without the hassle of public transport. Also supports:... It is an AI Agent Skill for Claude Code / OpenClaw, with 73 downloads so far.
How do I install Private Car?
Run "/install private-car" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Private Car free?
Yes, Private Car is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Private Car support?
Private Car is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Private Car?
It is built and maintained by xiejinsong (@xiejinsong); the current version is v3.2.0.
More Skills