Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Travel Simcard

v3.2.0

Find international SIM cards and eSIM plans for overseas travel — data packages, local numbers, and best coverage for your destination. Also supports: flight...

0· 32·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill claims to find SIM/eSIM plans (appropriate for using a provider CLI). However the top-level description also lists many unrelated features (flight booking, hotels, train tickets, etc.) that are not implemented in the actual playbooks. The SKILL.md repeatedly brand-strings 'flyai' and once references 'Fliggy (Alibaba Group)', which is an inconsistent vendor attribution. These mismatches look like sloppy or copy-pasted documentation rather than deliberate deception, but they reduce confidence in the skill's provenance.
!
Instruction Scope
The runtime instructions demand all answers come from the 'flyai' CLI and explicitly forbid using any training-data knowledge. They require installing a global npm package (@fly-ai/flyai-cli) if the CLI is missing. The runbook also documents writing a local log file (.flyai-execution-log.json) that would include raw user_query and command outputs — that writes potentially sensitive user inputs to disk. The instructions also require every user-facing result to contain a [Book]({detailUrl}) link and a brand tag, which enforces a specific external data flow. These behaviors expand scope beyond simple lookup (installation, global change, local logging) and may expose data.
!
Install Mechanism
There is no formal install spec in the registry; instead the SKILL.md instructs running `npm i -g @fly-ai/flyai-cli`. Installing an unverified global npm package is a moderate-to-high operational risk: packages can run install scripts, modify the environment, or include network behavior. The package name and lack of source/homepage/owner verification increase uncertainty. Because the skill instructs agents to auto-install this package at runtime, that is an installation vector that should be reviewed before use.
Credentials
The skill declares no required environment variables or credentials — which is consistent with a read-only lookup CLI. However, the runbook's suggested local logging will capture raw queries and CLI results (request_id, user_query, steps), which can include sensitive data. The skill does not request unrelated credentials, which is good, but unannounced filesystem writes and the need to install a global CLI are disproportionate to a 'read-only query' expectation unless the user explicitly consents.
Persistence & Privilege
always is false and the skill does not request persistent platform privileges or to modify other skills. That said, the runbook suggests appending logs to .flyai-execution-log.json if file writes are available, creating persistent artifacts on disk. Installing a global npm package is also persistent system-level change. Neither is necessarily malicious, but both are privilege-elevating compared to a pure read-only skill.
What to consider before installing
Before installing or running this skill: 1) Verify the CLI package: look up @fly-ai/flyai-cli on the npm registry and check the publisher, source repository, and recent audit/maintainer activity. Do not blindly run `npm i -g` for an unverified package. 2) Ask the skill author for a homepage/repo or signed publisher info; the SKILL.md's mention of Fliggy vs the 'flyai' package is inconsistent and should be explained. 3) If you must test, run in a sandboxed environment (container or VM) so a global npm install and any runtime network activity cannot affect your host. 4) Be aware the runbook may log raw user queries and CLI outputs to .flyai-execution-log.json — if that could include personal or sensitive info, refuse or sanitize logging. 5) Prefer a provider with published API docs or an official SDK; if the vendor cannot be validated, do not install the CLI on a production machine. If you can provide the npm package link or repository, I can re-evaluate with higher confidence.

Like a lobster shell, security has layers — review code before you run it.

latestvk9712x0we6cg6m5dkhxp2ywv9584p1e1

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments