The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell commands to manage Docker containers and create the project structure (e.g., docker ps, docker-compose up, mkdir -p). These are standard initialization steps for the described book project and occur within the local environment.
[DATA_EXFILTRATION]: The skill reads the .env file to check if API keys are set. This is a local configuration check used solely to warn the user if setup is incomplete; no data is transmitted externally or to unauthorized domains.
[EXTERNAL_DOWNLOADS]: No remote content is downloaded or executed. The skill runs a local script located within the project's own directory structure (./scripts/init-schema.sh) and uses local templates.
[PROMPT_INJECTION]: The skill ingests user input via $ARGUMENTS (Ingestion point: SKILL.md) which is formatted into a markdown header (Boundary markers: Markdown bolding). The skill possesses shell execution capabilities (Capability inventory: Bash, Write). However, the input is only used for project documentation and is not interpolated into executable commands (Sanitization: Not required for non-executable display).

bookstrap-init