redesigning-from-first-principles
Redesigns an existing design as if a new requirement had been a founding assumption, then propagates it to every reference and delivers it in working increments. Use when the user explicitly asks to redesign for a requirement or from first principles. Never infer from a feature request. Not for redoing an attempt; use redoing-implementations.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 2
- Forks
- 0
01 前言 Front Matter
- name
- redesigning-from-first-principles
- description
- Redesigns an existing design as if a new requirement had been a founding assumption, then propagates it to every reference and delivers it in working increments. Use when the user explicitly asks to redesign for a requirement or from first principles. Never infer from a feature request. Not for redoing an attempt; use redoing-implementations.
- disable-model-invocation
- true
02 资源文件 1
- 参考文档agents/openai.yaml327 B
03 SKILL.md 正文
Redesign From First Principles
Do not bolt the new requirement onto the current design. Redesign as if it had been a founding assumption, so the result reads as if it was always built that way.
- Read the whole design. Read every file the requirement touches and the files that depend on them, until you can state the current design's assumptions. Name the assumption the requirement breaks.
- Design from scratch. Ask: if this were written today with the requirement known from day one, what would be built? Describe that target design, then the bolt-on alternative, and say what the bolt-on would leave behind (special cases, flags, adapters, names that no longer fit). If the target design is the current design plus a small addition, say so and make that change; never redesign for its own sake.
- Plan the increments. Break the path to the target into steps that each leave the system working and its checks passing. Order them so no step needs a later one.
- Get approval. Reply with the broken assumption, the target design, what the bolt-on would leave behind, and the increments. Wait for approval before editing.
- Deliver one increment at a time. After each, run the project's checks and fix until they pass before starting the next.
- Propagate to every reference. Search for the broken assumption by name and by idea: types, names, comments, docs, examples, tests, fixtures, config, and rationale or decision sections. Update each to the new design, so none still describes the old one. Report what you changed and any reference you left, with why.
04 同仓库更多 Skills
writing-system-prompts
Writes and reviews system prompts for AI tools, agents, chatbots, and LLM products using a fixed section order and a quality checklist. Use when the user says "write a system prompt", or when drafting a system prompt or auditing one for missing stop criteria, tool rules, constraints, or examples. Not for compressing an existing prompt; use improving-prompts.
writing-prose
Writes and edits prose for people in a plain, conversational, short house style with no em dashes. Use when the user says "write this up", or when drafting or revising documentation, READMEs, explanations, blog posts, or other human-facing prose. Not for prompts written for agents; use writing-system-prompts or composing-agent-prompts.
writing-acceptance-criteria
Writes or rewrites acceptance criteria until each is provable: one observable outcome with a named check that can fail. Use when the user says "write acceptance criteria", or asks to write, rewrite, or tighten acceptance criteria or a definition of done. Not for proving work meets them; use proving-claims.
summarizing-shipped-work
Summarizes the user's merged GitHub PRs for a timeframe into a grouped, linked shipped-work report with highlights, and offers to update a brag document. Use when the user says "what shipped?", or when asked what the user shipped (this week, last month, a quarter, since a date), for a status update, or to update a brag doc.
running-retros
Mines a session transcript, past agent sessions, or named sources such as PR review comments for durable learnings, then proposes approval-gated edits to skills, AGENTS.md, or docs and tracker issues. Use when the user says "retro", or when the user explicitly asks for a retrospective. Never infer from session end or friction.
learning-from-mistakes
Codifies the latest correction or mistake in the conversation as a short imperative rule in the project or global CLAUDE.md or AGENTS.md, after the user approves it. Use when the user says "learn from this mistake", or when the user asks to learn from a mistake, remember a lesson for next time, add a rule to CLAUDE.md, or not repeat a behavior. Not for a whole session; use running-retros.