← Back to Skills Marketplace
32
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install opc-sandbox-simulation
Description
基于4个商业决策实战案例提炼的沙盘推演工具，帮助OPC创业者在重大决策前模拟多种场景，预测结果，降低试错成本。核心：先推演再行动。
README (SKILL.md)
胡田-OPC导师-沙盘推演
技能描述
基于4个商业决策实战案例（复用率44%）提炼的沙盘推演工具，帮助OPC创业者在重大决策前模拟多种场景，预测结果，降低试错成本。核心：先推演再行动。
触发条件
- 用户面临重大商业决策
- 用户需要评估多种方案的优劣
- 用户想预判决策的连锁反应
- 用户需要压力测试商业计划
核心能力
1. 沙盘推演五步法
- 定义决策：明确要做什么决定？选项有哪些？
- 设定变量：哪些因素会影响结果？变化范围？
- 构建场景：乐观/中性/悲观三个基准场景
- 推演链路：每个场景的一阶、二阶、三阶效应
- 决策建议：综合三个场景的最优选择+止损线
2. 场景构建框架
| 场景 | 概率设定 | 关键假设 | 预期结果 |
|---|---|---|---|
| 乐观 | 20-30% | 核心假设全部成立 | 最好情况 |
| 中性 | 40-50% | 部分假设成立 | 最可能情况 |
| 悲观 | 20-30% | 核心假设多数不成立 | 最差情况 |
3. 连锁效应分析
- 一阶效应：决策直接产生的结果
- 二阶效应：一阶效应引发的变化
- 三阶效应：二阶效应引发的系统性影响
- 关键：90%的决策失败源于忽略二阶、三阶效应
4. 止损机制设计
- 预设止损线：亏损达到X% / 时间超过Y个月
- 触发条件：关键指标连续3个月低于预期
- 退出方案：提前设计的退出路径
实战案例（4个验证）
| 案例 | 决策类型 | 推演场景数 | 关键发现 |
|---|---|---|---|
| 求实智源融资 | 融资方案选择 | 3 | 沙盘推演发现A轮估值过高风险 |
| GMF市场进入 | 跨境电商定价 | 3 | 3万元定价在悲观场景下亏损 |
| HNB合同签署 | 合同条款谈判 | 3 | 沙盘发现隐性成本被低估 |
| 商业计划书验证 | BP可行性 | 3 | 沙盘推演暴露收入假设过于乐观 |
OPC专属建议
- 一人公司每次重大投入前必须沙盘推演
- 重点关注：现金流断裂场景（悲观情景下的生存期）
- 推演深度：至少到二阶效应
- 用AI做沙盘推演，10分钟完成传统需要1周的分析
输出格式
- 三场景推演报告（乐观/中性/悲观）
- 连锁效应分析图
- 风险-收益矩阵
- 止损机制设计方案
Usage Guidance
Installers should be aware that this skill may engage in more general decision-making conversations than expected. If installed, consider narrowing its trigger to explicit scenario-simulation or decision-review requests.
Capability Assessment
Purpose & Capability
The described capability is a decision or scenario-simulation framework, which is coherent with helping users evaluate options or major choices.
Instruction Scope
The trigger language appears broad enough to activate during ordinary decision discussions; this is a usability and scoping issue, not evidence of malicious behavior.
Install Mechanism
No artifact-backed install script, binary payload, package mutation, or privileged setup behavior was identified; VirusTotal telemetry reported no malicious or suspicious engine hits.
Credentials
Available evidence does not show requests for filesystem crawling, credentials, browser sessions, network authority, or unrelated local access.
Persistence & Privilege
No persistence, background worker, privilege escalation, credential storage, or long-running agent behavior is evidenced.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install opc-sandbox-simulation - After installation, invoke the skill by name or use
/opc-sandbox-simulation - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
opc-sandbox-simulation 1.0.0
- 沙盘推演五步法（定义→变量→场景→推演→建议）
- 乐观/中性/悲观三场景构建
- 连锁效应分析（一阶/二阶/三阶）
- 基于OPC平台生态三会治理体系
Metadata
Frequently Asked Questions
What is 胡田-OPC导师-沙盘推演?
基于4个商业决策实战案例提炼的沙盘推演工具，帮助OPC创业者在重大决策前模拟多种场景，预测结果，降低试错成本。核心：先推演再行动。 It is an AI Agent Skill for Claude Code / OpenClaw, with 32 downloads so far.
How do I install 胡田-OPC导师-沙盘推演?
Run "/install opc-sandbox-simulation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 胡田-OPC导师-沙盘推演 free?
Yes, 胡田-OPC导师-沙盘推演 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 胡田-OPC导师-沙盘推演 support?
胡田-OPC导师-沙盘推演 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 胡田-OPC导师-沙盘推演?
It is built and maintained by golngod (@golngod); the current version is v1.0.0.
More Skills