bootstrap

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs at runtime to clone and run the installer from github.com/pedroromeroluna/ai-first-os (run /core/install.sh), which fetches and executes remote code required for the skill to function.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill tells the agent/operator to clone a GitHub repo and run an install script that "hooks" session contracts and creates symlinks (i.e. modifies system state and potentially system files), which can require elevated privileges or otherwise change the machine environment.