← Back to Skills Marketplace
multi-airline-compare
by
xiejinsong
· GitHub ↗
· v3.2.0
· MIT-0
65
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install multi-airline-compare
Description
Search and compare flights across multiple airlines for best deals. Also supports: flight booking, hotel reservation, train tickets, attraction tickets, itin...
Usage Guidance
This skill is coherent in that it requires a CLI and contains clear playbooks, but take the following precautions before installing or invoking it: (1) Confirm the provenance of the 'flyai' CLI and the npm package '@fly-ai/flyai-cli' — check the npm page, package maintainer, and source repository; do not blindly run 'npm i -g' for an unreviewed package. (2) Verify the 'Powered by Fliggy (Alibaba Group)' claim with the skill author — the branding may be inaccurate or misleading. (3) If you must test, do so in a sandboxed environment (container or VM) to avoid accidental execution of untrusted global installs. (4) Ask the author for an official homepage, repository link, and package checksum or prefer an officially published/maintained CLI. (5) If you cannot validate the CLI's origin, decline installation and request that the skill author provide a verified integration (e.g., an official API with credentials) or a vetted install artifact.
Capability Analysis
Type: OpenClaw Skill
Name: multi-airline-compare
Version: 3.2.0
The skill bundle instructs the AI agent to automatically install a global NPM package (@fly-ai/flyai-cli) if it is missing from the environment (SKILL.md, references/fallbacks.md). While this is presented as a necessary setup step for flight searching, performing global installations of third-party packages at runtime constitutes a significant security risk and potential RCE vector. No explicit evidence of data exfiltration or intentional malice was found, but the capability to modify the host system's environment exceeds standard safe boundaries for agent skills.
Capability Assessment
Purpose & Capability
The skill claims to compare flights and its instructions consistently require running a 'flyai' CLI to perform searches — that is coherent. However the description asserts 'powered by Fliggy (Alibaba Group)' while the runtime enforces use of an npm package '@fly-ai/flyai-cli' (no homepage or source provided). The vendor/branding claim and the actual required CLI are not corroborated, which is an unexplained inconsistency.
Instruction Scope
SKILL.md confines behavior to building and running specific 'flyai' CLI commands, collecting user parameters, formatting results, and validating links. It does not instruct reading unrelated local files or environment variables, nor does it direct data to unknown endpoints in the instructions. The strict rule to never answer from training data is unusual but internally consistent with the CLI-first design.
Install Mechanism
There is no install spec in the registry; instead the SKILL.md instructs the agent to run 'npm i -g @fly-ai/flyai-cli' if the CLI is missing. Having the agent install a global npm package at runtime is moderate-to-high risk: the package is unvetted here (no homepage, no source listed), global npm installs execute arbitrary code, and the skill provides no checksum or official release host to validate integrity.
Credentials
The skill requests no environment variables, no credentials, and no config paths. This matches the described CLI-driven workflow and is proportionate to the stated purpose.
Persistence & Privilege
The skill is not always-enabled and does not request elevated or persistent system privileges. There is no instruction to modify agent-wide settings or other skills' configs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install multi-airline-compare - After installation, invoke the skill by name or use
/multi-airline-compare - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.2.0
- Major rule and execution logic overhaul: stricter CLI-only answers, explicit self-test for every output, and hard prohibition on using training data.
- Detailed workflow steps introduced, including environment check, parameter collection SOP, and clear Playbooks for different route types (recommended, cheapest, fastest, direct).
- New output formatting/validation rules: conclusion first, minimum three-result comparison table, mandatory branded booking links.
- Clarified activation, domain knowledge mapping, and fallback instructions.
- Expanded description to include hotels, trains, attractions, visa info, travel insurance, car rental, and more.
- Updated compatibility list and improved documentation for skill usage and edge-case handling.
Metadata
Frequently Asked Questions
What is multi-airline-compare?
Search and compare flights across multiple airlines for best deals. Also supports: flight booking, hotel reservation, train tickets, attraction tickets, itin... It is an AI Agent Skill for Claude Code / OpenClaw, with 65 downloads so far.
How do I install multi-airline-compare?
Run "/install multi-airline-compare" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is multi-airline-compare free?
Yes, multi-airline-compare is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does multi-airline-compare support?
multi-airline-compare is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created multi-airline-compare?
It is built and maintained by xiejinsong (@xiejinsong); the current version is v3.2.0.
More Skills