Back to skill
Skillv1.0.0
ClawScan security
CurlShip Directory Submission · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 11, 2026, 3:44 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is an instruction-only wrapper describing simple curl calls to curlship.com and requests no credentials or installs — its claims and runtime instructions are internally consistent.
- Guidance
- This skill simply tells the agent to make HTTPS calls to https://curlship.com and requires you to provide a product URL and contact email — no credentials or installs are needed. Before installing: (1) verify you trust curlship.com (you're sending URLs and an email to an external service); (2) avoid submitting internal/private URLs — the doc claims SSRF protection but you should not rely on it for sensitive targets; (3) if you follow an upgrade workflow, inspect the checkout URL domain before entering payment details; (4) because it’s instruction-only, the main risk is data exfiltration of any URLs/emails you pass to it, so test with non-sensitive data first and review the service's privacy/terms.
Review Dimensions
- Purpose & Capability
- okThe name/description (submit a product to CurlShip) matches the SKILL.md: all required actions are HTTPS calls to curlship endpoints. No unrelated credentials, binaries, or installs are requested.
- Instruction Scope
- okSKILL.md only instructs the agent to POST/GET JSON to the documented API endpoints and to return responses; it explicitly requires a product URL and an email. There are no instructions to read local files, environment variables, or other system state, nor to transmit data to unexpected endpoints.
- Install Mechanism
- okThere is no install spec and no code files — this is instruction-only. That minimizes on-disk persistence and reduces installation risk.
- Credentials
- okThe skill declares no environment variables, no credentials, and the documented API is public/no-auth. The SKILL.md does instruct sending a user-provided email and URL to the external service, which is coherent with the stated purpose.
- Persistence & Privilege
- okThe skill does not request always:true and does not modify agent or system settings. Autonomous invocation is allowed by platform default but is not elevated here and is proportionate to the skill's functionality.