yu-iskw

Executes coding, review, research, and automation tasks using Google Antigravity CLI (`agy`). Uses Antigravity's native fine-grained allow/ask/deny permissions instead of Gemini CLI approval modes.

Executes tasks using the Claude Code CLI (`claude`). Automatically determines the safest permission mode (plan, acceptEdits, auto, or bypassPermissions) based on the task type.

Executes tasks using the Cursor Agent CLI (`cursor-agent`). Provides structured modes for planning, Q&A, and full automation with safety guardrails.

Implements new Agent Skills for the project. Identifies the AI coding tool (Cursor, Claude Code, Codex, Pi, Antigravity), ensures specification compliance, and provides specialized templates. Use when creating, authoring, or adding a new skill, or when the user asks about Agent Skills format or SKILL.md.

Executes tasks using the Codex CLI (`codex`). Automatically determines the least privilege required (Suggest, Auto-Edit, or Full-Auto) based on the user's request and handles security approvals.

Executes tasks using the GitHub Copilot CLI (`copilot`). Automatically determines the safest permission mode (plan, editor, or autopilot) based on the task type.

Executes coding, review, research, and automation tasks using the Vercel Labs fx coding-agent CLI (`fx`). Uses fx's native permission review, Agent Skills support, and single-request `fx ask` workflow while requiring external containment for untrusted or unattended work.

Executes coding, review, research, and automation tasks using the Pi coding agent CLI (`pi`). Uses explicit tool allowlists to apply least privilege and distinguishes project trust from sandboxing.