authenticated-fetch

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). The bundle downloads and executes a remote shell script directly from a raw.githubusercontent.com URL (a supply‑chain risk that can distribute malware if the GitHub content or account is compromised), while the other entries are either local health checks, benign example API endpoints, or malformed/wildcard placeholders.

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

• Third-party content exposure detected (high risk: 0.85). Yes: the required workflow fetches a user-supplied URL and then returns the raw response body (scripts/fetch_url.sh:183-193), which can include outsider-authored free text from arbitrary public web content into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's bundled setup wrapper (scripts/setup.sh) downloads a shell script from https://raw.githubusercontent.com/FinpeakInc/frevana-cli-releases/refs/heads/main/skills/frevana/scripts/setup.sh at runtime via curl and then execs it with bash, meaning remote code is fetched and executed as a required dependency.