linkfox-temu-manage-product-global

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill includes examples and CLI usage that embed API keys/accessToken directly in command-line JSON arguments (and shows exporting a key), which would require the LLM to output secret values verbatim and thus risks exfiltration.

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

• Secret detected (high risk: 1.00). I scanned the skill docs and scripts for literal, high-entropy credentials. Most occurrences are placeholders or examples (e.g., "TOKEN", "", "PASTE_TOKEN", environment variable names like LINKFOXAGENT_API_KEY, masked tokens shown by list_stores, or short/obvious example passwords) which per the rules are documentation placeholders or setup/example values and are ignored.

One actual literal stands out: a base64url-looking UID value assigned to _LOGIN_FIXED_UID in scripts/onboarding.py. It is not an obvious placeholder (it is a long, encoded string), is used as a fixed header value in the login flow (comment: "登录链路固定 uid header（实测可用）") and therefore could be a real token/credential that is functional in requests. Because it is a long encoded token (high-entropy-looking) and used to construct request headers for login, I classify it as a potential secret leak.

All other suspect strings in the repo are either clearly placeholders (TOKEN, "", "PASTE_TOKEN", examples in JSON/bash blocks), environment variable names, or masked/derivative values and are therefore ignored per the provided rules.