The Agent Skills Directory

[COMMAND_EXECUTION]: The skill generates shell scripts that utilize gh (GitHub CLI) for secret management, browser opening utilities (xdg-open, open, explorer.exe), and standard file operations. These commands are integral to the skill's stated purpose of automating repository setup.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill handles sensitive data, such as API keys and secrets. It employs security best practices by using hidden input prompts (ask_secret) and persisting values only to local .env files or the user's own GitHub repository via the official CLI. No unauthorized remote transmission was detected.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze repository files to determine the necessary setup steps, creating a potential ingestion surface for instructions embedded in project documentation.
Ingestion points: Reads local files including README, .env, docker-compose.yml, and GitHub workflow files (SKILL.md).
Boundary markers: None specified for the ingested content.
Capability inventory: File system writes (.env), GitHub secret configuration (gh CLI), and shell command execution (template.sh).
Sanitization: Not explicitly mentioned for the data read from project files.

wizard