linkfox-seerfar-ozon-market-keyword-search

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text via the user-supplied keywords/matchType fields passed into the runtime API call POST /seerfar/ozon/marketKeywordSearch (and then prints/saves the returned keyword objects), so an outsider can submit arbitrary text into the query payload that the system processes at runtime.

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

• Secret detected (high risk: 1.00). I searched for high-entropy literal values that could function as real credentials. I found a base64url-looking string hardcoded as the default for _LOGIN_FIXED_UID in scripts/onboarding.py (assigned when the LINKFOX_LOGIN_FIXED_UID env var is absent). The code comment says this fixed UID header is "实测可用" (tested usable) and it is used as the uid header when no user_id is present, so it can act as a functional authentication header. This is a high-entropy literal (not a placeholder or simple setup password) and therefore should be treated as a secret.