linkfox-zhihuiya-description-data-translated

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). Skill runtime path scripts/zhihuiya_description_translated.py calls POST /zhihuiya/descriptionDataTranslated using user-supplied patentId/patentNumber, then ingests and prints/summarizes the API response field data[].description (translated patent text) into stdout/context; the externally authored text originates from the patent database records selected by the user.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). I scanned the skill files for literal, non-placeholder high-entropy credentials. The only candidate that meets the "high-entropy, literal value used to access a service" criteria is the base64url string assigned to _LOGIN_FIXED_UID in scripts/onboarding.py. It is not a placeholder name (it's a concrete encoded string), is used as a UID header in requests, and the inline comment indicates it is actually usable ("实测可用"), so it appears to be a real, fixed token/header value rather than an example. No other high-entropy API keys, private keys, or secrets are present; other values are environment variable names, example IDs, or obvious non-secret strings (ignored per the rules).