← Back to Skills Marketplace
sucriss

Decision Framework

by Criss_Su · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
276
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install decision-framework
Description
决策框架分析技能。提供多种经典决策模型（SWOT、10-10-10 法则、决策树、加权打分、Eisenhower 矩阵）帮助用户结构化思考。Use when user asks for help making a decision, analyzing options, evaluating choices, o...
README (SKILL.md)

Decision Framework - 决策框架

Overview

本技能提供 5 种经典决策框架，帮助用户结构化分析选择、评估方案、做出理性决策。每个框架输出结构化分析 + 建议倾向（但不替用户做最终决定）。

支持的框架

1. SWOT 分析

适用场景： 战略决策、职业选择、项目评估、竞争分析

结构：

  • S (Strengths) - 内部优势
  • W (Weaknesses) - 内部劣势
  • O (Opportunities) - 外部机会
  • T (Threats) - 外部威胁

交互示例：

用户：我在纠结要不要辞职创业，帮我 SWOT 分析
你：好的，咱们从四个维度分析：
     【优势】你有哪些核心竞争力？...
     【劣势】目前有什么短板？...
     【机会】市场/行业有什么机会？...
     【威胁】可能面临什么风险？...

输出： 四象限列表 + 关键洞察 + 建议倾向

2. 10-10-10 法则

适用场景： 情绪化决策、短期冲动 vs 长期价值、人际关系决策

结构：

  • 10 分钟后 - 即时感受/后果
  • 10 个月后 - 中期影响
  • 10 年后 - 长期意义

交互示例：

用户：要不要答应这个聚会邀请（其实不太想去）
你：用 10-10-10 法则看看：
     10 分钟后：可能有点累但当下开心
     10 个月后：这个聚会对你的生活有什么影响？
     10 年后：这个决定还重要吗？

输出： 三时间维度分析 + 情绪 vs 理性对比 + 建议

3. 决策树分析

适用场景： 多阶段决策、有风险的选择、需要计算期望值的场景

结构：

  • 根节点：核心决策
  • 分支：不同选择
  • 子分支：可能的结果 + 概率 + 收益/损失

交互示例：

用户：要不要投资这个创业项目
你：咱们画个决策树：
     投资 → 成功 (30%, +500 万) / 失败 (70%, -100 万)
     不投资 → 维持现状 (100%, 0)
     期望值计算：投资 = 0.3×500 + 0.7×(-100) = 80 万

输出： 树状结构 + 期望值计算 + 风险收益比 + 建议

参考： references/decision-tree-guide.md 查看详细决策树模板

4. 加权打分法 (Weighted Scoring)

适用场景： 多选项对比、需要量化评估的决策、采购/选型

结构：

  1. 列出评估维度（如：成本、质量、时间、风险）
  2. 给每个维度分配权重（总和 100%）
  3. 每个选项在各维度打分（1-10 分）
  4. 计算加权总分

交互示例：

用户：三个 offer 选哪个
你：先确定评估维度和权重：
     薪资 (30%)、成长 (25%)、工作强度 (20%)、地点 (15%)、团队 (10%)
     
     然后给每个 offer 打分：
     Offer A: 薪资 8×0.3 + 成长 7×0.25 + ... = 7.2 分
     Offer B: ...
     Offer C: ...

输出： 维度权重表 + 各选项得分表 + 排名 + 建议

脚本： scripts/weighted-scoring.py 自动计算加权分数

5. Eisenhower 矩阵（优先级矩阵）

适用场景： 任务优先级排序、时间管理、资源分配

结构：

  • 紧急 + 重要 → 立即做
  • 重要 + 不紧急 → 计划做
  • 紧急 + 不重要 → 委托做
  • 不紧急 + 不重要 → 不做/少做

交互示例：

用户：手头一堆事不知道先做哪个
你：咱们用 Eisenhower 矩阵分类：
     【紧急重要】今天必须交的報告、紧急 bug 修复
     【重要不紧急】学习新技能、锻炼身体、长期项目
     【紧急不重要】某些会议、临时打扰
     【不紧急不重要】刷社交媒体、无意义娱乐

输出： 四象限分类 + 行动建议 + 时间分配建议

工作流程

Step 1: 理解决策场景

询问用户：

  • 你在纠结什么？（核心决策）
  • 有哪些选项？（A/B/C 或 是/否）
  • 最关心什么因素？（成本/时间/风险/收益/...）

Step 2: 推荐或直接选择框架

主动推荐逻辑：

  • 战略/职业/项目 → SWOT
  • 情绪化/冲动决策 → 10-10-10
  • 有风险/多阶段 → 决策树
  • 多选项对比 → 加权打分
  • 任务优先级 → Eisenhower 矩阵

或让用户选择：

想用哪个框架分析？
1. SWOT - 优势/劣势/机会/威胁
2. 10-10-10 - 短期/中期/长期视角
3. 决策树 - 风险 + 期望值计算
4. 加权打分 - 多选项量化对比
5. Eisenhower 矩阵 - 任务优先级
6. 你帮我推荐

Step 3: 收集信息

根据选择的框架，引导用户提供必要信息：

  • SWOT → 询问内外部因素
  • 10-10-10 → 询问各时间维度的考量
  • 决策树 → 询问选项、概率、收益/损失
  • 加权打分 → 询问维度、权重、各选项表现
  • Eisenhower → 询问任务列表

Step 4: 输出分析

结构化输出：

  • 框架分析结果（列表/表格/树状图）
  • 关键洞察（2-3 条）
  • 建议倾向（但不替用户决定）
  • 可选：导出 markdown 保存

Step 5: 可选深化

  • 需要更详细分析某个维度吗？
  • 要对比其他框架的结论吗？
  • 需要导出保存吗？

框架选择指南

决策类型 推荐框架 原因
职业选择 SWOT + 10-10-10 全面评估 + 长期视角
投资决策 决策树 + SWOT 风险量化 + 内外部分析
多选项对比 加权打分 量化排名
情绪化决策 10-10-10 跳出当下情绪
任务优先级 Eisenhower 紧急/重要分类
战略决策 SWOT + 决策树 全局 + 风险

输出格式规范

Markdown 表格（加权打分示例）

| 维度 | 权重 | Offer A | Offer B | Offer C |
|------|------|---------|---------|---------|
| 薪资 | 30%  | 8 (2.4) | 9 (2.7) | 7 (2.1) |
| 成长 | 25%  | 7 (1.75)| 6 (1.5) | 9 (2.25)|
| ...  | ...  | ...     | ...     | ...     |
| **总分** | 100% | **7.2** | **7.5** | **7.8** |

树状图（决策树示例）

是否投资创业？
├── 投资 (投入 100 万)
│   ├── 成功 (30%) → 收益 500 万
│   └── 失败 (70%) → 损失 100 万
│   期望值：0.3×500 + 0.7×(-100) = 80 万
└── 不投资
    └── 维持现状 (100%) → 0

四象限（SWOT/Eisenhower）

【优势 Strengths】          【劣势 Weaknesses】
- 点 1                     - 点 1
- 点 2                     - 点 2

【机会 Opportunities】       【威胁 Threats】
- 点 1                     - 点 1
- 点 2                     - 点 2

注意事项

  1. 不替用户做决定 - 提供分析和建议倾向，但最终决定权在用户
  2. 保持客观 - 避免带入主观偏见，基于用户提供的信息分析
  3. 量化优先 - 能数字化的尽量数字化（概率、分数、权重）
  4. 承认不确定性 - 明确标注假设和不确定因素
  5. 可导出保存 - 支持导出 markdown 供用户后续参考

Resources

scripts/

  • weighted-scoring.py - 加权打分自动计算脚本

references/

  • decision-tree-guide.md - 决策树详细模板和示例
  • frameworks-comparison.md - 各框架对比和适用场景详解
Usage Guidance
This skill appears coherent and implements what it claims: decision frameworks plus a small local Python helper for weighted scoring. Before installing, consider: (1) The included Python script runs locally — review it (done here) and it's a simple calculator that only processes command-line inputs. (2) The skill will handle whatever decision text you provide, so avoid pasting highly sensitive credentials or secrets into prompts. (3) If you plan to let the agent execute the script automatically, run it in a sandbox or environment you control (standard caution). If additional files were added that perform network calls, downloads, or request credentials, re-evaluate because that would change the assessment.
Capability Analysis
Type: OpenClaw Skill Name: decision-framework Version: 1.0.0 The decision-framework skill bundle is a legitimate tool designed to help users analyze choices using structured models like SWOT and Decision Trees. The included Python script, `scripts/weighted-scoring.py`, is a simple utility for calculating weighted averages and contains no dangerous functions, network calls, or file system interactions. All instructions in `SKILL.md` and documentation in the `references/` directory are aligned with the stated purpose and show no signs of malicious intent or prompt injection.
Capability Assessment
✓ Purpose & Capability
The name/description (decision frameworks: SWOT, 10-10-10, decision tree, weighted scoring, Eisenhower) align with the SKILL.md and included files. The single script (weighted-scoring.py) directly supports the '加权打分' feature. There are no declared environment variables, binaries, or config paths that are unrelated to the stated purpose.
✓ Instruction Scope
SKILL.md instructs the agent to ask the user for decision context, gather options/weights/probabilities, and produce structured outputs (tables, trees, insights). It references only local resources (the included script and reference docs). There are no instructions to read arbitrary system files, access environment variables, or send data to external endpoints.
✓ Install Mechanism
No install specification is provided (instruction-only-style). The repository contains one small, local Python script; there are no downloads, external package installs, or extracted archives. Because nothing is fetched from the network during install, risk is low.
✓ Credentials
The skill declares no required environment variables or credentials. The SKILL.md and code do not reference secrets or unrelated services. The Python script accepts only command-line inputs (dimensions/options) and produces a markdown table — no credential access or external integrations are present.
✓ Persistence & Privilege
always:false and normal agent invocation settings. The skill does not request permanent presence, does not modify other skills' configs, and does not ask to store credentials. No elevated privileges are requested.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install decision-framework
  3. After installation, invoke the skill by name or use /decision-framework
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of decision-framework skill: - Provides structured decision-making support using five classic frameworks: SWOT, 10-10-10 rule, Decision Tree, Weighted Scoring, and Eisenhower Matrix. - Helps users analyze situations, compare options, and prioritize tasks with clear, step-by-step guidance and structured outputs. - Suggests frameworks based on scenario keywords and user needs (e.g., "纠结", "要不要", "SWOT", "weighted scoring" etc.). - Outputs include structured analysis, key insights, and recommendation tendencies—without making decisions for users. - Supports markdown export for analysis results and includes scripts/resources for deeper analysis.
Metadata
Slug decision-framework
Version 1.0.0
License MIT-0
All-time Installs 2
Active Installs 2
Total Versions 1
Frequently Asked Questions

What is Decision Framework?

决策框架分析技能。提供多种经典决策模型（SWOT、10-10-10 法则、决策树、加权打分、Eisenhower 矩阵）帮助用户结构化思考。Use when user asks for help making a decision, analyzing options, evaluating choices, o... It is an AI Agent Skill for Claude Code / OpenClaw, with 276 downloads so far.

How do I install Decision Framework?

Run "/install decision-framework" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Decision Framework free?

Yes, Decision Framework is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Decision Framework support?

Decision Framework is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Decision Framework?

It is built and maintained by Criss_Su (@sucriss); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments