Back to skill
Skillv1.0.0

ClawScan security

JD.com Product Comments API · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 29, 2026, 4:07 AM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
The skill's code, runtime instructions, and required environment variables align with its stated purpose (calling JustOneAPI's JD.com comments endpoint) and do not request unrelated credentials or perform surprising actions.
Guidance
This skill appears coherent with its purpose, but review these practical points before installing: (1) The JUST_ONE_API_TOKEN is sent as a query parameter by the helper — tokens in URLs can be exposed in logs, referer headers, or CLI history; avoid sharing or pasting the token and consider using a token with limited scope/rotation. (2) The helper prints raw JSON to stdout; be careful if you pipe or log outputs as they may contain sensitive data. (3) Verify you trust JustOneAPI (https://api.justoneapi.com) and that the token you provide has only the minimal necessary permissions. (4) Ensure your agent environment has a recent Node installed (fetch is used).

Review Dimensions

Purpose & Capability
okName/description match the code and SKILL.md: the skill calls JustOneAPI's GET /api/jd/get-item-comments/v1. Required runtime (node) and the single env var JUST_ONE_API_TOKEN are consistent with this purpose.
Instruction Scope
noteSKILL.md and bin/run.mjs instruct running a local node helper that performs an HTTP GET to the documented JustOneAPI endpoint and prints raw JSON. One notable detail: the API token is passed as a query parameter (operation parameter named 'token'), which can expose the token in logs or referer headers — this is an API design choice, not hidden malicious behavior. The skill requests only the itemId and optional page; it does not read other files or env vars.
Install Mechanism
okNo install script; this is instruction-only with an included small helper script (bin/run.mjs). No downloads from external URLs or package installs are performed by the skill itself. The only runtime requirement is the node binary.
Credentials
okOnly JUST_ONE_API_TOKEN is required and declared as the primary credential, which matches the documented API usage. There are no unrelated or excessive environment variables or config path requirements.
Persistence & Privilege
okalways is false and the skill does not request persistent system-wide configuration or modify other skills. It can be invoked autonomously (platform default), which is expected for skills of this type and is not combined with other red flags.