Gpt Series Reasoning Style

Process-discipline layer only — not a reasoning-capability booster and not GPT-specific despite the legacy name. Use when coordinating multiple independent models or agents (commander / subagent / single modes), or when building any deliverable that needs structured, high-rigor execution — pre-implementation gate, resource survey, role identities, 23-field task packages, DRI closure, trust tiers (T1/T2/T3), evidence verification, hands-on UX verification, and final acceptance. Triggers: 实现前确认, 任务包, 多Agent协作, 指挥官模式, 资源盘点, 实操验收, UNVERIFIED.

JadeYingWah Updated

File contents

JadeYingWah/gpt-series-reasoning-style/tree/main/ commit 6bad121e04

Frequently asked questions

npx skillmds@latest add jadeyingwah/gpt-series-reasoning-style