Back to skill
Skillv1.0.1

ClawScan security

China Express Query · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 13, 2026, 10:32 PM
Verdict
Benign
Confidence
high
Model
gpt-5-mini
Summary
The skill's code, instructions, and requirements are coherent with a simple courier-tracking tool that queries third‑party public APIs (kuaidi100 / Baidu); nothing requests unrelated credentials or system access.
Guidance
This skill appears to do exactly what it says: a Node CLI that queries public courier tracking APIs. Before installing, consider that it will make outbound HTTP requests to third‑party services (kuaidi100 and a Baidu endpoint) and send whatever tracking numbers you provide — if those numbers are sensitive, avoid passing them or run the script locally. No credentials are requested. If you need stricter guarantees, review the small script (scripts/query.mjs) yourself; it is concise and easy to audit.

Review Dimensions

Purpose & Capability
okName/description match the implementation: a Node CLI that detects courier company from a tracking number and queries public tracking APIs. Required binary is only node and no unrelated secrets, which is proportionate to the stated purpose.
Instruction Scope
okSKILL.md instructs running the provided script with a tracking number and optional flags. The runtime behavior is limited to network queries to kuaidi100 and a Baidu query endpoint, formatting output, and optionally writing the output to a file. The instructions do not ask the agent to read unrelated files, environment variables, or to transmit unrelated data.
Install Mechanism
okNo install spec — instruction-only with a small included script. This has low install risk; nothing is downloaded from arbitrary URLs or added to system paths.
Credentials
okThe skill requests no environment variables or credentials. The only external interaction is HTTP GETs to public APIs for tracking info, which is appropriate for this functionality.
Persistence & Privilege
okThe skill is not always-enabled, is user-invocable, and does not modify other skills or system-wide settings. It only writes an output file if the user supplies --output.