The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the official GitHub CLI (gh) to perform repository operations. The instructions provide templates for common tasks like viewing PRs, merging, and querying the GitHub API.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from GitHub (issues, pull requests, and comments), which is a common attack surface for indirect prompt injection.
Ingestion points: Remote content is ingested via gh pr view, gh issue list, and gh api commands within SKILL.md.
Boundary markers: The skill does not explicitly define delimiters for external data, but uses --jq to select specific fields from the JSON responses.
Capability inventory: The agent has access to the gh CLI tool and local shell execution for command templates.
Sanitization: No specific instruction-sanitization is present beyond structured field selection via jq.

github