The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute setup scripts from external sources.
Evidence: curl -fsSL https://get.docker.com -o get-docker.sh followed by sudo sh get-docker.sh in SKILL.md for Docker installation.
Evidence: curl -fsSLo "$tmpdir/homebrew-install.sh" https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh for Homebrew installation.
Evidence: curl -fsSLo "$tmpdir/nodesource-setup.sh" https://deb.nodesource.com/setup_20.x followed by execution via sudo -E bash for Node.js setup.
Note: These sources are official distribution channels for the respective tools, which are well-known technology services.
[COMMAND_EXECUTION]: The skill recommends using sudo for package management and environment configuration tasks.
Evidence: Usage of sudo apt update, sudo apt install, and sudo usermod for Linux environment configuration.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to execute commands (like npm install or pip install) within repositories cloned from user-provided URLs.
Ingestion points: The agent clones external repositories using git clone based on user input.
Boundary markers: None identified in the provided instructions.
Capability inventory: The skill performs package installations (npm install, pip install) and shell script execution.
Sanitization: No explicit validation of the cloned repository's contents is performed before installation commands are run.

environment-setup-guide