This skill dispatches five distinct developer persona subagents (Theo Browne, DHH, Tanner Linsley, ThePrimeagen, Pieter Levels) in parallel. Each persona answers the user's technical question within stated constraints using a pros/cons/confidence (/pcc) format. The skill then validates outputs against a definition-of-done and synthesizes agreement, disagreement, and the strongest take.
Use this when you want multiple, opinionated perspectives on architecture, tooling, or trade-offs—for example: choosing a stack, comparing approaches, or getting diverse code-review viewpoints. Not for step-by-step implementation; focuses on WHY over HOW.
Best with agents that support parallel subagent dispatch and structured outputs (Claude Code, Cursor, Codex-style systems).
This skill has not been reviewed by our automated audit pipeline yet.