The Agent Skills Directory

[SAFE] (SAFE): The skill does not contain any malicious patterns or unauthorized data access commands. It uses the standard gh CLI tool for its intended purpose of repository management.
[Indirect Prompt Injection] (SAFE): Although the skill ingests PR data (titles, branch names) from GitHub, it uses this data in a non-executable manner (displaying to user and extracting PR numbers) for CLI arguments, which minimizes the surface area for indirect injection.
[Command Execution] (SAFE): Command execution is restricted to the gh tool as defined in the allowed-tools metadata, preventing arbitrary command injection.

merge-stack