update-services

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content contains explicit credential-theft guidance (asking for the user's root/sudo password and instructing to echo it into sudo), executes an unsigned remote install script via curl | sh (classic supply-chain risk), and provides instructions to enable wide network/CORS exposure and unsanitized environment injection into a systemd service — together these are high-risk, easily abused patterns for credential exfiltration, remote compromise, and persistent backdoor installation.

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

• Third-party content exposure detected (high risk: 0.90). The skill includes a required runtime script (scripts/install_ollama.sh) that runs curl -fsSL https://ollama.com/install.sh | sh, directly fetching and executing a public third‑party webpage/script (ollama.com) which can materially change behavior and system state.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill invokes and executes remote code at runtime via "curl -fsSL https://ollama.com/install.sh | sh" (in scripts/install_ollama.sh) and pulls+runs a remote container image "ghcr.io/open-webui/open-webui:main" (in scripts/update_opencode.sh), both of which fetch external code that is executed and are required by the skill.