基于 SOC 职业分类
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/benjam3n/reasoningtool --skill saqrc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
Route any input through a branching question tree to narrow down the optimal response strategy before writing. Two stages — PERCEIVE (classify input) then ACT (select response). Covers all prompt types.
Generate exhaustive guesses about user input using ALL search methods with coverage tracking. Guessing is SEARCH through possibility space. Tracks space created vs space covered to ensure comprehensive exploration.
Systematically evaluate and select from a set of guesses, options, or possibilities. Combines ARAW analysis with prioritization to determine which guesses are strong, weak, actionable, or eliminable.
| name | saqrc |
| description | Rewrite unclear questions by separating intent from target question. |
| output | {"format":"prose"} |
Input: $ARGUMENTS
Rewrite unclear questions by separating intent ("what did I mean?") from the target question ("what do I want to answer?"). Many questions in gates and procedures are answerable only if you already know what the author meant — which defeats the purpose.
Extract question-like lines from the target. Prioritize:
For each candidate, apply the Stranger Test: Could a competent stranger answer this question the same way as the author would, without additional interpretation?
| Question | Stranger Test | Failure Mode |
|---|---|---|
| [question] | Pass/Fail | [what's unclear] |
For each failing question, iterate:
REWRITE CHAIN:
Original: [the question as written]
Round 1:
What's unclear: [what requires interpretation]
Intent: [what did the author MEAN to ask]
Rewrite: [clearer version]
Round 2 (if still unclear):
What's still unclear: [remaining ambiguity]
Intent: [what THAT was supposed to mean]
Rewrite: [even clearer version]
Round 3 (if needed):
...
Final: [question that can be answered without interpretation]
Rewrite principles:
For each final rewrite:
QUESTION REWRITE AUDIT:
Target: [what was analyzed]
Questions examined: [N]
Failed stranger test: [N]
Rewrites produced: [N]
Top rewrites:
| Original | Rewrite | Why Better |
|----------|---------|-----------|
| [unclear] | [clear] | [what changed] |
Patterns found: [common clarity issues across questions]