Back to skill
Skillv1.0.0

ClawScan security

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

Scanner verdict

ReviewApr 6, 2026, 8:04 PM
Verdict
Review
Confidence
medium
Model
gpt-5-mini
Summary
The skill's stated purpose (fetching ZenQuotes) is simple, but the runtime instructions direct the agent to call a third‑party Pipeworx gateway and to bootstrap remote code via npx — actions that are not clearly justified by the description and introduce moderate risk.
Guidance
This skill could be benign (a simple quotes proxy) but the README asks you to contact a remote Pipeworx gateway and to bootstrap remote code with npx. Before installing or running it, consider: 1) Verify who operates https://gateway.pipeworx.io and whether you trust that operator to see any data the agent sends; 2) Avoid running npx mcp-remote@latest on a privileged host — inspect the package source first or run it in a sandbox; 3) If you prefer a simpler setup, ask the author why the gateway is needed instead of calling the public ZenQuotes API directly; and 4) If you will route any sensitive context through the agent while this skill is enabled, do not enable it until you confirm the data-handling policy of the remote service.

Review Dimensions

Purpose & Capability
noteName and description promise quotes from ZenQuotes, which is plausible. However, the SKILL.md points all traffic at gateway.pipeworx.io rather than directly to a ZenQuotes API; this is a minor mismatch that could be legitimate (a proxy/service) but is not explained in the README.
Instruction Scope
concernThe examples instruct POSTing to https://gateway.pipeworx.io/zenquotes/mcp and the Setup block instructs adding an mcpServers entry that launches npx mcp-remote@latest pointed at that gateway. That directs the agent to connect to and execute commands against a remote MCP endpoint and to bootstrap remote code via npx — behavior beyond merely fetching quotes and which can transmit arbitrary request/response data to the external gateway.
Install Mechanism
concernThe package has no formal install spec, but the SKILL.md tells users to run npx mcp-remote@latest, which will download and execute code from npm at runtime. Using npx to pull the latest remote package is a moderate-to-high risk action because it runs remote code that isn't vetted by this skill bundle.
Credentials
okNo environment variables, secrets, or config paths are requested by the skill metadata. The lack of requested credentials is proportionate to a read-only quotes service.
Persistence & Privilege
okalways is false and there are no indications the skill demands persistent system-wide privileges or to modify other skills. The skill will be invoked normally and does not claim to auto-enable itself.