菜单
Forces discipline on every non-trivial request: restate the user's intent, diagnose the root cause, name the domain and reuse existing mature solutions before improvising, then deliver one chosen answer instead of a menu of options.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/recomby-ai/promptly-prompt --skill promptly-prompt命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | promptly-prompt |
| description | Forces discipline on every non-trivial request: restate the user's intent, diagnose the root cause, name the domain and reuse existing mature solutions before improvising, then deliver one chosen answer instead of a menu of options. |
AI answer quality depends on context quality, not prompt tricks.
This skill operates via a UserPromptSubmit hook that injects two
disciplines into complex requests:
Restate, then pause if needed — echo the user's request in your own words, surface implicit constraints, name what would make the answer wrong. If anything is ambiguous or your planned approach might not be endorsed, stop and ask before acting. If the restatement makes it clear there is no disagreement, continue. Skip the restatement only when the request is genuinely trivial.
Super-dimensional view — before solving from memory, name the domain, search for established methodologies, frameworks, libraries, and prior art, then bring that specialist knowledge to bear. Cite specific names, not vague gestures. If nothing fits, say so and explain why.
The hook script at scripts/intercept.py scores prompt complexity using
rule-based signals. Simple commands pass through untouched. Complex requests
get the full injection.
When invoked directly (e.g., user says "optimize this prompt"), apply the two disciplines manually: restate the user's intent with implicit needs surfaced, then locate the domain and the existing methods that belong to it. Rewrite the prompt with that context filled in.