linkfox-ai-mode-google-search

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md + scripts/google_ai_search.py: the skill takes user-supplied keyword, sends it as q= to Google AI Mode, and then ingests the resulting Google AI Overview Markdown (stdout/result) back into the agent for summarization/output.

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 values are environment-variable names, example placeholders, or short example IDs (which per rules are ignored). However, scripts/onboarding.py defines a non-obvious base64-like literal string as a fixed UID header default:
• _LOGIN_FIXED_UID = ... "eyJhX2lkIjoiNmEyMmM4YjA1YmM5MTZhIiwiZF9pZCI6IiJ9"

This is a hardcoded, random-looking token (base64url JSON) used as a uid header when no access token is supplied. Because it is a literal high-entropy value that the code treats as a usable header in authentication flows, it meets the definition of a secret and should be treated as sensitive.

No other high-entropy API keys, private keys, or credentials were found. Many other strings are environment-variable names, instructions, or examples and are intentionally ignored.