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 rsg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 | rsg |
| description | Handler for goals focused on REBUILDING from damage, not improving from baseline. |
| output | {"format":"prose"} |
Input: $ARGUMENTS
Handler for goals focused on REBUILDING from damage, not improving from baseline. These are distinct because: you're not building on a foundation — you're rebuilding FROM damage, there's often an emotional/psychological recovery component, path dependence matters, trust/credibility may need to be re-earned, and the "before" state may not be achievable or desirable.
| Priority | Criterion | Action |
|---|---|---|
| Emergency | Damage is still happening | STOP THE BLEEDING first |
| Urgent | Will get worse without action | Address within days |
| Important | Won't get worse but blocks recovery | Address within weeks |
| Long-term | Full restoration | Address over months |
Rule: Stabilize before rebuilding. Don't start constructing while the ground is still shaking.
Recovery isn't just practical — it's emotional:
Not all recovery goals have an emotional component, but many do. Ignoring it sabotages the practical recovery.
The "before" state is often not the right target:
| Option | When It Applies |
|---|---|
| Restore: Return to previous state | Damage was external, previous state was good |
| Rebuild better: Use lessons to improve | Damage revealed weaknesses worth fixing |
| Adapt: Accept new reality, build something different | Previous state is impossible or undesirable |
| Reframe: What felt like damage is actually an opportunity | Sometimes true, but don't force it |
Choose the target explicitly. Don't default to "back to how it was."
Different from a standard plan because of path dependence:
RECOVERY PLAN:
Phase 1 — Stabilize (now):
- [stop ongoing damage]
- [secure what's left]
Phase 2 — Quick Recovery (days-weeks):
- [easy wins that restore confidence]
- [rebuild basic functionality]
Phase 3 — Deep Recovery (weeks-months):
- [address root causes]
- [rebuild trust/credibility/capacity]
Phase 4 — Growth (months+):
- [build beyond previous state]
- [apply lessons learned]
Cannot recover: [what must be accepted as permanent]
If the damage involved others' trust:
Use the recovery to become stronger than before:
RECOVERY PLAN:
Damage: [what was lost/broken]
Cause: [root cause]
Current state: [honest baseline]
Recovery target: [restore / rebuild better / adapt / reframe]
Recovery timeline:
Phase 1 (stabilize): [actions] — by [when]
Phase 2 (quick wins): [actions] — by [when]
Phase 3 (deep recovery): [actions] — by [when]
Phase 4 (growth): [actions] — by [when]
Cannot recover: [permanent losses accepted]
Trust repair: [if applicable — specific steps]
Anti-fragility: [how to be stronger after recovery]
Emotional status: [acknowledged / processing / resolved]