菜单
Use when starting creative work, planning a multi-step task, isolating a workspace, wrapping a branch, or giving/getting code review. Not when the task is a one-line fix with no design surface.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Supreme policy layer governing all Claude Code behavior. Autonomy, one-line prompt interpretation, speed standards, emphasis signal processing, cross-skill coordination, done definitions, conflict resolution. Loaded every prompt.
Establish project thesis before first code. Infer product type from domain/folder/README. Identify users, business model, pSEO strategy, AI-native dev approach. Maintain PROJECT_BRIEF.md as source of truth.
Deep web research, competitor scanning, technology evaluation, and implementation planning. Decomposes work into vertical slices, identifies parallel workstreams, tracks assumptions with confidence levels, and designs the critical path for minimum wall-clock time.
Captures and evolves user preferences with confidence levels. Maintains Voice of the Customer model with exact language, dissatisfaction and aspiration signals. Handles promotion/demotion, global vs project scoping, auto memory system, and Omi wearable data integration.
Cloudflare-first platform selection. Decision trees for Workers, D1, R2, KV, DO, Queues, Vectorize, Containers, Sandboxes, Flagship, Agent Memory, Workflows v2. Default stack, override conditions, auth, data patterns, reliability.
Implements features in vertical slices, always starting with homepage. Enforces anti-placeholder rules — no lorem ipsum, no TODO stubs, no gray boxes. Real content, real images, real interactions. TypeScript strict mode, Zod validation, and structured file organization.
| name | superpowers |
| description | Use when starting creative work, planning a multi-step task, isolating a workspace, wrapping a branch, or giving/getting code review. Not when the task is a one-line fix with no design surface. |
| when_to_use | Process skills that govern HOW to approach work — invoke BEFORE implementation skills. Brainstorm before building; plan before coding; worktree before isolated work; review before merge; finish-branch when done. |
| effort | high |
| model | inherit |
| priority | 2 |
| pack | core |
| stage | stable |
| triggers | ["brainstorm","let's build","write a plan","implementation plan","git worktree","code review","review my code","finish branch","write a skill","subagent driven"] |
| paths | ["*"] |
| submodules | ["brainstorming/SKILL.md","writing-plans/SKILL.md","subagent-driven-development/SKILL.md","using-git-worktrees/SKILL.md","finishing-a-development-branch/SKILL.md","requesting-code-review/SKILL.md","receiving-code-review/SKILL.md","writing-skills/SKILL.md"] |
Vendored from obra/Superpowers (MIT, Jesse Vincent), compressed to house style — see NOTICE.md. These are process skills: they decide HOW to approach a task and run BEFORE implementation skills. On conflict, this repo's 01-operating-system + rules/* + Brian's preferences win.
brainstorming/SKILL.md — explore intent + design before code. Non-negotiable first step.writing-plans/SKILL.md — write the plan before touching code.using-git-worktrees/SKILL.md — spin an isolated workspace ([[main-only-branch]]).subagent-driven-development/SKILL.md — one subagent per task + two-stage review ([[monitor-orchestration]], [[parallel-subagent-economy]]).requesting-code-review/SKILL.md — dispatch the code-reviewer agent + Agent Diversity Review ([[agent-selection]]).receiving-code-review/SKILL.md — verify it technically, then act; no blind agreement.finishing-a-development-branch/SKILL.md — merge / PR / cleanup ([[no-staging-doctrine]], auto-push).Authoring a skill itself? → writing-skills/SKILL.md + [[skill-authoring-contract]] (the house authoring rule, where the SDO / match-the-form / ordered-by-weight insights now live).
requesting-code-review → the code-reviewer agent + diversity gate.subagent-driven-development → meta-orchestrator / parallel Agent spawns.brainstorming → pairs with 14-independent-idea-engine.test-driven-development → [[e2e-tdd-organization]]systematic-debugging → [[error-recovery]]verification-before-completion → [[verification-loop]]dispatching-parallel-agents → [[parallel-subagent-economy]]executing-plans → [[monitor-orchestration]]using-superpowers → 01-operating-system (skill-check before action)[[skill-authoring-contract]]