The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of shell commands and piping, including the use of jq, base64, and diff. It also documents gh alias set --shell, which allows users to define shell expressions that are evaluated through the sh interpreter when the alias is invoked.
[REMOTE_CODE_EXECUTION]: Documentation for gh extension install is provided, which allows the installation of third-party CLI extensions that execute locally. The skill also describes gh workflow run for triggering GitHub Actions workflows.
[DATA_EXFILTRATION]: The skill contains commands to fetch and decode remote file content using the GitHub API (gh api ... contents) and to display authentication tokens (gh auth token or gh auth status --show-token), which could lead to sensitive data exposure if used on untrusted resources.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from remote repositories.
Ingestion points: File contents, directory listings, and repository metadata are fetched via gh api in SKILL.md and references/remote-analysis.md.
Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the fetched repository content.
Capability inventory: The skill has broad capabilities, including shell command execution, network requests via the GitHub CLI, and the ability to install and run external extensions.
Sanitization: No sanitization, escaping, or validation of the retrieved repository content is performed before processing.

gh-cli