linkfox-google-patent-search

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

• Third-party content exposure detected (low risk: 0.10). Google Patents Search ingests only the user-supplied search query string q (and other parameters) to call POST /googlePatent/search, and the required workflow is a list-style search over first-party/publisher content rather than reading arbitrary outsider-authored text from an email/chat/ticket/feed/issue body.

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

• Secret detected (high risk: 1.00). I scanned all files for literal, high-entropy values that could be usable credentials. Most values are environment-variable names, URLs, user-agent strings, or clear placeholders (e.g., references to LINKFOX_AGENT_API_KEY, ${LINKFOX_TOOL_GATEWAY}, curl examples) which per the rules are ignored.

However, scripts/onboarding.py contains a hardcoded base64url-like string assigned as a default UID header:

• _LOGIN_FIXED_UID default = "eyJhX2lkIjoiNmEyMmM4YjA1YmM5MTZhIiwiZF9pZCI6IiJ9"

This is not a placeholder (it is a concrete, high-entropy value) and is used as a header when building auth requests if no user_id is present. Because it appears to be an actual token-like value (base64url JSON/JWT segment) embedded in code and used for authentication headers, it meets the definition of a secret and should be flagged.

Other values (env var names, example curl uses, simple strings) were ignored as documentation placeholders or non-sensitive defaults per the provided rules.