The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple gh CLI commands to verify authentication and fetch project details, fields, and items. These are standard operations for the tool's purpose.
[PROMPT_INJECTION]: There is an indirect prompt injection surface as the skill processes project metadata from GitHub into local JSON files.
Ingestion points: GitHub project metadata and item lists fetched via the gh CLI.
Boundary markers: The skill relies on structured JSON output but does not explicitly define sanitization steps for the metadata values.
Capability inventory: Bash (specifically for gh and mkdir), Read, and Write operations.
Sanitization: None explicitly mentioned for the data fetched from the GitHub API.

ghpm-init