x-topic-search

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The raw.githubusercontent.com link is a direct setup shell script from an external/unverified GitHub repo that the wrapper downloads and executes (running code directly from raw GitHub is a high-risk distribution vector), while the 127.0.0.1:${DAEMON_PORT}/health endpoint is a benign local health check — overall the package is suspicious due to the remote setup script.

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

• Third-party content exposure detected (high risk: 0.75). The required runtime path runs scripts/setup.sh, which downloads and executes a remote script from https://raw.githubusercontent.com/.../setup.sh; that outsider-authored free text (the downloaded shell script) is executed and can influence what the agent/LLM context receives indirectly (indirect prompt injection risk).

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 downloads and execs remote code at runtime from https://raw.githubusercontent.com/FinpeakInc/frevana-cli-releases/refs/heads/main/skills/frevana/scripts/setup.sh, so the skill fetches and executes external script content it depends on.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill requires running a bundled setup script that downloads and executes a remote installer (and may install/start a local daemon or binaries), which modifies the machine state and can execute arbitrary code and may require elevated privileges.