Back to skill
Skillv1.0.0
ClawScan security
Ship · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 8, 2026, 5:25 PM
- Verdict
- benign
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's instructions and requirements are coherent with a logistics/shipping advisor: it is instruction-only, requests no credentials or installs, and stays on-topic — but provenance is unknown and the skill will expect/shuffle sensitive shipment data when used, so exercise usual caution.
- Guidance
- This skill appears coherent for shipping/logistics advice and doesn't request installs or credentials in the manifest, but review the full SKILL.md before installing because it may ask you to provide sensitive shipment data or carrier API keys during use. Before sharing any real customer addresses, HS codes, or invoice values: (1) confirm the skill's source/author (unknown here); (2) avoid bulk pasting of customer lists in one go — test with dummy/example shipments first; (3) if you connect carrier accounts, use scoped/rotating API keys or service accounts you can revoke; (4) review any instructions that mention external endpoints (APIs or webhooks) to ensure they target legitimate carrier domains; (5) limit the agent's autonomous actions if you don't want it to act on credentials or submit claims without confirmation.
Review Dimensions
- Purpose & Capability
- okThe name and description (shipping/logistics intelligence) match the SKILL.md content (classification of scenarios, carrier comparison, international rules, rate-shopping pseudocode). There are no unrelated environment variables, binaries, or install steps declared that would be out-of-scope for this purpose.
- Instruction Scope
- noteThe instructions are focused on shipping workflows and include pseudocode and checklists. They will require user-supplied shipment data (addresses, HS codes, declared values, etc.), which is expected for this domain. I found no directives in the visible SKILL.md to read arbitrary system files, environment variables, or send data to unexpected endpoints. Caveat: because the file is long and partially truncated in the prompt, I cannot confirm there are no sections that recommend contacting external services or requesting API keys — those would be expected but should be reviewed for exact endpoints and data flows.
- Install Mechanism
- okNo install spec and no code files — instruction-only skill introduces minimal install risk (nothing written to disk by the skill itself).
- Credentials
- okThe manifest declares no required environment variables or credentials. That aligns with the instruction-only nature. Note: practical use (automated rate shopping or carrier account integration) would typically require carrier API keys or platform credentials; the skill does not request those up-front, so any credential exchange would happen interactively and should be scrutinized by the user.
- Persistence & Privilege
- okalways:false and no special OS or config-path access. The skill can be invoked autonomously by the agent (platform default), which is normal for skills; nothing indicates permanent or elevated privileges.