The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use shell commands including git and gh (GitHub CLI) for repository initialization and management. These operations are preceded by environment verification (e.g., checking authentication status) and multiple layers of user confirmation to prevent unauthorized actions.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface related to processing local directory data, which could contain untrusted instructions.
Ingestion points: The agent scans the current directory's file structure and paths to determine which files to publish.
Boundary markers: The skill requires the agent to provide a full summary of intended actions and file lists, requiring explicit user approval before any local or remote mutations occur.
Capability inventory: The skill possesses the ability to execute shell commands (git, gh) and write to the local file system (creating a README.md).
Sanitization: The instructions explicitly mandate the exclusion of credential files, .env variants, and suspicious paths to ensure sensitive data is not accidentally staged or pushed.

gh-this