linkfox-amazon-alexa-search

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow reads outsider-authored free text from the user-supplied prompts[0] in the /amazon/alexaSearch request body (and may include user-provided url), which is sent into the Alexa-backed response generation.

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

• Secret detected (high risk: 1.00). I flagged a hardcoded, literal token-like value present in the code. Specifically, scripts/onboarding.py defines _LOGIN_FIXED_UID as a base64url/JWT-like string used as a default UID header ("实测可用"). This is a high-entropy literal embedded in source that can be sent to upstream services and therefore meets the definition of a secret (literal value that may provide access).

I did NOT flag other items because they are placeholders or non-sensitive:

• Environment variable names (LINKFOX_AGENT_API_KEY / LINKFOXAGENT_API_KEY) are just variable names, not values.
• Example taskId / sample response values in README files are sample data, not credentials.
• URLs, user-agent strings, and other configuration defaults are not secrets per the rules.