The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read various repository files to identify necessary environment variables and secrets.
Ingestion points: The agent is directed to read .env, .env.example, README, docker-compose*, framework configurations, and GitHub workflow files in SKILL.md.
Boundary markers: There are no explicit instructions for the agent to treat this data as untrusted or to use delimiters when processing it.
Capability inventory: The agent generates executable bash scripts capable of writing to the local filesystem (.env files) and executing GitHub CLI commands (gh secret set).
Sanitization: No sanitization or validation steps are prescribed for the content extracted from repository files before it is used to generate the script stages.
[COMMAND_EXECUTION]: The generated wizard script utilizes shell commands to perform its tasks.
Evidence: The template.sh library includes functions that call xdg-open, open, wslview, and explorer.exe to open URLs, as well as gh secret set and gh variable set to interact with GitHub Actions. These are standard operations for the skill's stated purpose of automating environment setup.

wizard