Skillv1.0.1

ClawScan security

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

BenignMar 10, 2026, 3:15 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only travel-planning guide for China; its files, instructions, and requirements are internally consistent and do not request elevated access or external credentials.
Guidance: This skill appears to be a harmless, self-contained travel planning guide (itineraries, checklists, and recommendations). Before installing: confirm what tools your agent already has (web browsing, booking APIs, payment connectors). If your agent can perform network actions or make bookings, be aware the high-level instructions could be used to trigger bookings or transmit personal documents (passports, visa details) — consider restricting the agent's booking/payment tools or requiring explicit user confirmation for any transaction or for sharing sensitive documents. Otherwise, the skill itself requests no credentials or installs and is coherent with its stated purpose.

Purpose & Capability: okName/description (China travel planning, bookings, family logistics) matches the included SKILL.md and reference documents; no unrelated binaries, env vars, or config paths are requested.
Instruction Scope: noteSKILL.md is a high-level workflow and travel reference material (itineraries, checklists, documents). It does not instruct the agent to read local files, access credentials, or call any specific external endpoints, but it is intentionally general — depending on the agent's toolset the skill could be used to perform real bookings or network actions if those tools exist.
Install Mechanism: okNo install spec and no code files; this is instruction-only so nothing will be written to disk or fetched during installation.
Credentials: okNo environment variables, credentials, or config paths are requested — proportional to a documentation/instruction skill.
Persistence & Privilege: notealways is false and the skill is user-invocable; model invocation is allowed (platform default). This is normal but means the agent could autonomously use the skill if the agent has tools that permit bookings or network access.