before-and-after

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Yes — while cli.github.com and gist.github.com are legitimate, the list includes anonymous/unmoderated file hosts and raw file endpoints (0x0.st, a custom blob upload URL, and gist.githubusercontent.com raw links) that can directly serve executables or scripts and are commonly used to distribute malware.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's required pre-flight step runs a runtime install of remote code via "npm install -g @vercel/before-and-after", which fetches and then executes a third-party CLI that the skill relies on for captures, so this is a runtime external dependency that executes remote code.

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

• Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill instructs running system commands that install a global npm package and to use the Vercel CLI (including obtaining a bypass token) and other local scripts—actions that modify the machine state and can be used to bypass deployment protections—though it does not explicitly request sudo, edit system files, or create users.