跳到正文
cccskills
免费在 GitHub 公开

shape

Decide the technical path once the ask is clear. Use when the WHAT is confirmed but more than one way to build it remains — architecture, technology, design.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(2)

  • SKILL.md1.9 KB
  • loop-selection.md879 B

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Route

Decide only among ways that differ where it matters. Where no work order states the bar the ask must meet, use deepen skill first. Look up what settles a choice without the user — the bar, their taste from CLAUDE.md and project rules, what their own accounts and plans already offer, facts from the repo, docs, the web. Only choices that stay open after that route to a skill.

Route each open choice by what keeps it open, invoking the named skill with the Skill tool:

  • A claim about something that exists — a library fits, the schema holds, the current design bears the load → use judge skill.
  • The user can't choose from nothing — nothing exists to react to → use draft skill: the fewest versions that show a real fork, differing where it matters — who owns the data, where the seam sits, what fails how.
  • The user holds taste they can't yet state → use elicit skill.
  • Their stated preference reads two ways that would build different things → use clarify skill.
  • They're missing something already settled about the technology → use explain skill.

When the work iterates — an agent loop, a refinement cycle — which loop is itself one of the choices: read loop-selection.md.

Each decision re-enters the routing: a resolved choice can open the next one or conflict with an earlier one; a conflict is one more choice, surfaced to the user. Route again until every open choice has the user's decision or confirmation and none conflict.

The path

Ask any choice still open first, with AskUserQuestion. Then state each resolved choice with its reason and what lost.

The ask itself is still rough — starting would mean inventing what the user means → use intent skill.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

anchor

免费

Rewrite a rule so anyone applies it the same way. Use when a stated rule or claim takes the reader's judgment to apply.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

clarify

免费

Turn words with two readings into one confirmed ask. Use when the user's words admit two readings that would build different things.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

complain

免费

Surfaces what the agent would change and kept to itself — a decision it would make differently, effort it saw wasted, a better route to the goal, a fact the user lacks, friction in the tools, permissions, setup or docs the work ran on. Use when you hold one of these and have not said it.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

compose

免费

Plan the skills an ask needs, invoke each as its condition arrives. Use when the ask has multiple aspects that different skills carry, or the user named multiple skills.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

consult

免费

Bring how named experts would think about the question. Use when the user wants expert thinking on a tradeoff, choice, or stuck problem — not a verdict.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

Have the agents behind rival answers to one question check each other's answers, then pick on the evidence. Use when independent answers to one question disagree, or would, and no test, file or run can settle which is right.

中文简介正在准备中。当前显示原文说明。

saadshahd/moo.md382026年10月22日 更新

查看 saadshahd 的全部技能

报告此技能的问题