The Agent Skills Directory

[SAFE]: The skill consists entirely of documentation files in Markdown format. No executable scripts, binaries, or automated tasks are included within the skill structure.
[SAFE]: Mentions of sensitive file paths (e.g., ~/.ssh/id_rsa, ~/.config/glab-cli/config.yml) and placeholders for credentials (e.g., glpat-xxx) are present as part of the command reference and usage examples. No hardcoded secrets or data exfiltration logic were identified.
[SAFE]: The documentation includes instructions for users to download and install the CLI from various sources, including official and community-maintained repositories. These are provided as informative text and do not involve remote code execution by the agent itself.
[SAFE]: Internal instructions to the agent (e.g., CRITICAL: grep references/ for keywords before answering) are used as prompt engineering techniques to guide the agent toward the provided documentation and do not attempt to bypass safety filters.

glab-docs