The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to create labels. The command execution is transparently documented and limited to standard repository management operations.
[DATA_EXPOSURE]: The skill correctly identifies the use of GITHUB_TOKEN and GH_TOKEN as the standard method for API authentication. It provides non-functional placeholders and encourages best practices for secret management by suggesting the use of environment variables.
[INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by processing user-supplied data for label names and descriptions which are later interpolated into commands and API payloads.
Ingestion points: User-provided inputs for name, description, and color parameters described in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill definitions.
Capability inventory: Subprocess execution through the gh CLI and network communication with the GitHub REST API.
Sanitization: The instructions do not define explicit validation or sanitization requirements for the processed user strings.

github-create-label