The Agent Skills Directory

[COMMAND_EXECUTION]: The skill primarily operates by executing shell commands via a custom CLI tool named foyer. While intended for project management, this interface requires the agent to construct and run shell strings.
[COMMAND_EXECUTION]: There is a potential risk of shell command injection if user-supplied parameters, such as the project slug or description, are not properly sanitized before being passed to the CLI commands.
[PROMPT_INJECTION]: The skill ingests untrusted data from two main sources: direct user input for project details and external content retrieved via foyer search or foyer activity context.
Ingestion points: foyer search and foyer activity context (referenced in SKILL.md).
Boundary markers: None explicitly defined to separate retrieved content from system instructions.
Capability inventory: File system modification, Git initialization, and GitHub repository creation (via foyer project init).
Sanitization: No explicit sanitization or validation of the retrieved text is mentioned before processing.

init-project