← 返回 Skills 市场
payne-openclaw

Decision Helper

作者 Payne-OpenClaw · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
118
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install decision-helper
功能描述
🎯 The Ultimate Decision Weapon | 终极决策武器 Stop overthinking. Start deciding. 停止过度思考,开始果断决策。 Your AI-powered decision-making system that cuts through noise, we...
使用说明 (SKILL.md)

Decision Helper | 决策助手

🎯 Decisions made simple. Clarity delivered instantly.

决策本应简单,清晰即刻呈现。


Why This Skill? | 为什么选它?

传统方式 Decision Helper
纠结几小时 ⚡ 30秒出框架
凭感觉乱选 📊 结构化分析
事后后悔 🎯 提前看清利弊
一个人扛 🧠 AI智囊团支持

Your secret weapon for life's crossroads. 人生十字路口的秘密武器。


Trigger | 触发策略

自动触发(高置信度)

当同时满足:

  1. 决策场景 - 涉及选择/取舍
  2. 情绪信号 - 纠结/不确定/害怕
  3. 重要性 - 影响较大

信号组合:

场景 + 情绪 框架
"A还是B" + 无情绪 ❌ 不触发(简单选择)
"A还是B" + "纠结" ✅ 利弊分析
"要不要" + "怕后悔" ✅ 10-10-10
"哪个好" + 无情绪 ⚠️ 轻量提示
重大决策词 + 任何犹豫 ✅ 直接分析

重大决策词: 辞职、买房、结婚、转行、投资、搬家、创业

轻量提示(中置信度)

"看起来你在做选择,需要我帮你分析一下吗?"

等用户确认后再用框架

不触发(低置信度)

  • 简单事实问答
  • 用户明确知道答案
  • 日常琐事选择

Frameworks | 五大决策框架

1. 利弊分析 ⚖️

场景: 二选一

A: ✅ 优点 → ❌ 缺点 → 净评估
B: ✅ 优点 → ❌ 缺点 → 净评估

关键差异: [一句话]
建议: [A/B/再想想]

2. 决策矩阵 📊

场景: 多维度比较

标准      权重   A    B    C
────────  ────   ──   ──   ──
[标准1]   [x%]   [分] [分] [分]
[标准2]   [x%]   [分] [分] [分]
───────────────────────────────
加权总分         [分] [分] [分]

胜出: [选项]

3. 最坏情况 😰

场景: 害怕风险

选A的最坏结果: [具体描述]
能承受吗? [是/否]

选B的最坏结果: [具体描述]  
能承受吗? [是/否]

结论: [选能承受的那个 / 都不选 / 降低风险再选]

4. 10-10-10 ⏰

场景: 纠结当下感受

10分钟后: [感觉]
10个月后: [感觉]
10年后:  [感觉]

视角: [长期/短期]
建议: [选长期不后悔的]

5. 可逆性测试 ↩️

场景: 犹豫不决

这个决定可逆吗? [是/否]

是 → 快速决定,错了再改
否 → 慢点决定,多想几步

Response Template | 回应模板

Step 1: 识别 (1句话)

"这是个[类型]决策,用[框架]分析:"

Step 2: 分析 (3-5句话)

[直接套用框架]

Step 3: 建议 (1句话)

"建议: [具体建议],因为[关键理由]。"

Step 4: 提醒 (可选)

"注意: [潜在风险]"


Examples | 示例

User: "我该换工作吗?"

Response:

这是个利弊权衡决策,用利弊分析:

现工作: ✅ 熟悉/稳定 ❌ 没成长
新工作: ✅ 机会/钱 ❌ 风险/适应

关键差异: 稳定 vs 成长

**建议: 如果新工作最坏结果能接受,就去。**因为年轻时不搏的成本 > 失败成本。

注意: 确认新工作的真实情况,别只看表面。


User: "买A车还是B车?"

Response:

多维度比较,用决策矩阵:

标准 权重 A车 B车
价格 30% 8 6
油耗 25% 6 8
空间 25% 9 7
维修 20% 7 7
加权 100% 7.5 6.9

胜出: A车,空间优势抵消了价格差。

但看你实际最在意什么?


Judgment Criteria | 判断标准

用决策助手的条件

IF (重大决策 OR 明显纠结) AND 用户没明确倾向 → 自动分析
IF 普通选择 AND 用户犹豫 → 轻量提示
IF 简单选择 OR 用户明确 → 不触发

判断示例

用户输入 分析 动作
"今天吃啥" 日常琐事 ❌ 不触发
"我该辞职吗" 重大+不确定 ✅ 自动分析
"A还是B好" 普通选择 ⚠️ "需要分析吗?"
"A还是B,好纠结" 普通+情绪 ✅ 自动分析
"买房还是租房" 重大+权衡 ✅ 自动分析

Principles | 核心原则

  1. : 30秒内给出框架
  2. : 不解释原理,直接套用
  3. : 一句话点出关键
  4. : 最后留一个问题给用户
  5. : 判断该不该用,比怎么用更重要

The Promise | 承诺

Every decision is a bet on your future. 每个决策都是对未来的赌注。

This skill helps you stack the odds in your favor. 这个技能帮你把胜算握在手中。


版本: 1.0.0 | 类型: Native Skill

安全使用建议
This skill is instruction-only and coherent with its decision-making purpose — it doesn't request credentials or perform installs. Two practical notes before installing: (1) the skill describes automatic triggering on "hesitation" or "life-changing" signals; review your agent's skill-autotrigger settings so it doesn't run at moments you don't expect, and (2) treat outputs as structured suggestions, not professional legal/financial advice — verify high-stakes decisions with domain experts. If you want further assurance, confirm the agent environment prevents skills from making network calls or persisting data.
功能分析
Type: OpenClaw Skill Name: decision-helper Version: 1.0.0 The 'decision-helper' skill is a purely prompt-based (native) tool designed to assist users with decision-making frameworks like Pros/Cons and Decision Matrices. Analysis of SKILL.md and package.json reveals no executable code, system commands, network requests, or instructions to exfiltrate sensitive data; it functions entirely as a conversational guide for the AI agent.
能力评估
Purpose & Capability
Name/description (decision-making) align with the content of SKILL.md. The skill requires no binaries, env vars, or config paths — appropriate for an instruction-only decision helper.
Instruction Scope
SKILL.md contains only decision frameworks, trigger rules, response templates and examples. It does not instruct the agent to read files, access environment variables, call external endpoints, or collect unrelated data.
Install Mechanism
No install spec and no code files beyond package.json and SKILL.md. Nothing is written to disk or downloaded during install — lowest-risk pattern for a skill.
Credentials
The skill requests no credentials or environment variables. There are no config paths or secrets required, which is proportionate to its stated functionality.
Persistence & Privilege
always is false and agent invocation is normal default. The SKILL.md describes auto-triggering heuristics, but the skill itself does not request elevated persistence or modify other skills/config.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install decision-helper
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /decision-helper 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
decision-helper 1.0.0 – Initial Release - Introduces an AI-powered decision-making system designed to simplify complex life choices. - Provides auto-trigger logic for high-stakes decisions (e.g., career, investments, relationships) and hesitation signals. - Features five structured decision frameworks: Pros & Cons, Decision Matrix, Worst-Case Test, 10-10-10, and Reversibility Test. - Includes clear criteria for when and how the skill activates, with sample responses and guidance templates. - Bilingual support for both English and Chinese users. - Delivers concise, framework-based recommendations in under 30 seconds.
元数据
Slug decision-helper
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Decision Helper 是什么?

🎯 The Ultimate Decision Weapon | 终极决策武器 Stop overthinking. Start deciding. 停止过度思考,开始果断决策。 Your AI-powered decision-making system that cuts through noise, we... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 118 次。

如何安装 Decision Helper?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install decision-helper」即可一键安装,无需额外配置。

Decision Helper 是免费的吗?

是的,Decision Helper 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Decision Helper 支持哪些平台?

Decision Helper 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Decision Helper?

由 Payne-OpenClaw(@payne-openclaw)开发并维护,当前版本 v1.0.0。

💬 留言讨论