The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the gh CLI and gh api to perform administrative operations on GitHub repositories and projects, including creating issues, managing project fields, and configuring access roles.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data such as issue titles and bodies, which are used as arguments in shell commands and API requests. This creates an indirect prompt injection surface if malicious content is ingested.
Ingestion points: Issue metadata (titles, bodies), project field values, and user-supplied strings for epics and tasks (e.g., in SKILL.md and references/project-setup.md).
Boundary markers: The instructions do not provide specific delimiters or ignore-instructions for handling external content.
Capability inventory: High-privilege write operations via gh issue create/edit, gh project item-edit, and gh api (REST/GraphQL), including the ability to modify repository access roles.
Sanitization: While the skill uses recommended gh CLI flags and GraphQL variables, the agent must remain vigilant when interpolating these values into shell environments to prevent command injection.

git-project