The Agent Skills Directory

[SAFE]: The skill implements a structured engineering review process. It uses standard development tools (gh, git, grep) to analyze the current project state.
[SAFE]: The allowed-tools section includes Bash and WebSearch, which are used for standard development tasks such as checking PR status via the GitHub CLI and researching architectural best practices. These tools are used within the scope of the skill's primary purpose.
[SAFE]: The skill uses AskUserQuestion as a core mechanism for interactive decision-making, ensuring that the AI agent does not make significant architectural or code changes without explicit human approval.
[SAFE]: The 'Security review' section (Section 2) correctly identifies and advises on common vulnerabilities like injection vectors, authorization flaws, and secrets management using standard security principles.
[SAFE]: The 'Review Context Detection' section uses gh pr view and gh repo view to determine the execution context, which is a standard procedure for development-focused AI agents.

plan-eng-review