superpower-router

Design-first flow. For brainstorming/analysis work, use Codex (gpt-5.4) for research; after approval hand off to plan-and-execute for Codex (gpt-5.3-codex) implementation.

Alias of brainstorm command; Codex gpt-5.4 during brainstorming/analysis, Codex gpt-5.3-codex during implementation

Execute plan with proactive dual-model Codex dispatch (gpt-5.3-codex for code, gpt-5.4 for analysis) and fallback controls

Create and/or execute plans with explicit dual-model Codex routing (gpt-5.3-codex for code, gpt-5.4 for analysis)

Research-only mode. Route logic/reasoning to Codex (gpt-5.4); route live web research to Claude WebSearch.

Create an implementation plan with explicit backend routing per task

Use this agent when a major project step has been completed and needs to be reviewed against the original plan and coding standards. Examples: <example>Context: The user is creating a code-review agent that should be called after a logical chunk of code is written. user: "I've finished implementing the user authentication system as outlined in step 3 of our plan" assistant: "Great work! Now let me use the code-reviewer agent to review the implementation against our plan and coding standards" <commentary>Since a major project step has been completed, use the code-reviewer agent to validate the work against the plan and identify any issues.</commentary></example> <example>Context: User has completed a significant feature implementation. user: "The API endpoints for the task management system are now complete - that covers step 2 from our architecture document" assistant: "Excellent! Let me have the code-reviewer agent examine this implementation to ensure it aligns with our plan and follows best practices" <commentary>A numbered step from the planning document has been completed, so the code-reviewer agent should review the work.</commentary></example>

You MUST use this before creative implementation work. Use Codex (gpt-5.4) during brainstorming/research, then hand off approved plans for Codex (gpt-5.3-codex) execution.

Use when completing tasks, requesting review, or receiving code review feedback — covers both requesting and responding to reviews with technical rigor

Use when implementation is complete and you need to verify work and decide how to integrate — covers verification-before-completion and branch finishing

Use when creating implementation plans and executing tasks with proactive dual-model Codex routing (gpt-5.3-codex for code, gpt-5.4 for analysis) and fail-closed behavior before Claude/Sonnet fallback

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes

Use when implementing any feature or bugfix, before writing implementation code

Use when starting any conversation — establishes skill usage rules and multi-agent routing

Use when creating new skills, editing existing skills, or verifying skills work before deployment

2 hooks across 1 event