amazon-product-info

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The Amazon product URLs and the localhost health check are benign, but the raw.githubusercontent.com URL is suspicious because the skill downloads and directly executes a remote shell script from a third‑party GitHub repo (a common malware distribution vector).

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill intentionally downloads and executes a remote setup script and then uses a local Chrome-extension-backed daemon to access a logged-in Chrome session (via frevana), which constitutes a high-risk supply-chain / remote-execution pattern and an access vector that could be abused to steal session data.

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

• Third-party content exposure detected (high risk: 0.75). Yes—runtime calls frevana_ask with prompt set to the user-supplied Amazon product URL, and the Frevana Chrome-session tool is expected to fetch/parse that page’s content (outsider-authored web page text) into the LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The bundled scripts/setup.sh is invoked at runtime and downloads+executes remote code from https://raw.githubusercontent.com/FinpeakInc/frevana-cli-releases/refs/heads/main/skills/frevana/scripts/setup.sh (curl -o then exec bash), so the fetched content is executed and is required for the skill to run.