← 返回 Skills 市场

Decision Advisor

Name: Decision Advisor
Author: banxian87

作者 banxian87 · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

114

总下载

当前安装

版本数

在 OpenClaw 中安装

/install decision-advisor

功能描述

Decision-making advisor using Tree of Thoughts for exploring options, analyzing trade-offs, and providing data-driven recommendations.

使用说明 (SKILL.md)

Decision Advisor

AI-powered decision advisor that uses Tree of Thoughts to explore multiple options, analyze trade-offs, and provide structured recommendations for complex decisions.

Use Cases

🎯 Technical Decisions

Technology stack selection
Architecture patterns
Build vs buy decisions
Tool evaluation

💼 Business Decisions

Product feature prioritization
Investment analysis
Strategic planning
Risk assessment

👥 Personal Decisions

Career choices
Education paths
Major purchases

Features

🌳 Option Exploration

Generate multiple alternatives
Consider non-obvious options
Explore different perspectives

⚖️ Trade-off Analysis

Pros and cons for each option
Risk assessment
Cost-benefit analysis
Long-term implications

📊 Structured Framework

Decision criteria definition
Weighted scoring
Sensitivity analysis
Recommendation with confidence level

Usage

const advisor = new DecisionAdvisor();

const decision = await advisor.advise({
  decision: 'Choose frontend framework for enterprise app',
  options: ['React', 'Vue', 'Angular'],
  criteria: [
    { name: 'Performance', weight: 0.3 },
    { name: 'Developer Experience', weight: 0.25 },
    { name: 'Ecosystem', weight: 0.25 },
    { name: 'Long-term Support', weight: 0.2 }
  ],
  context: 'Team of 20 developers, 5-year project lifespan'
});

console.log(decision.recommendation);
console.log(decision.analysis);

Example Output

## Decision Analysis: Frontend Framework Selection

### Options Evaluated
1. React
2. Vue
3. Angular

### Scoring Matrix

| Criteria              | Weight | React | Vue | Angular |
|-----------------------|--------|-------|-----|---------|
| Performance           | 30%    | 8.5   | 8.0 | 7.5     |
| Developer Experience  | 25%    | 9.0   | 8.5 | 7.0     |
| Ecosystem             | 25%    | 9.5   | 7.5 | 8.5     |
| Long-term Support     | 20%    | 9.0   | 7.0 | 9.0     |
| **Weighted Score**    |        | **8.95** | 7.85 | 7.95 |

### Recommendation

**React** is recommended with a weighted score of 8.95/10.

### Key Reasons
1. Largest ecosystem and community
2. Excellent developer experience
3. Strong corporate backing (Meta)
4. Proven at scale

### Risks & Mitigations
- **Risk**: Frequent breaking changes
- **Mitigation**: Use stable LTS versions, invest in testing

### Confidence Level: High (85%)

Architecture

Decision Request
    ↓
Tree of Thoughts Agent
    ├─ Generate options
    ├─ Explore variations
    └─ Identify criteria
    ↓
Evaluation Phase
    ├─ Score each option
    ├─ Analyze trade-offs
    └─ Assess risks
    ↓
Recommendation Engine
    ├─ Weighted scoring
    ├─ Sensitivity analysis
    └─ Confidence calculation
    ↓
Structured Report

Installation

clawhub install decision-advisor

License

MIT

Version

1.0.0

Created

2026-04-02

安全使用建议

This skill appears coherent and light-weight. Two practical things to consider before installing: (1) It expects the host to provide an LLM implementation (this.llm.generate). The included default LLM simply returns '5' so you should supply a real LLM client (and the credentials that client requires) to get useful results. (2) Any sensitive context you pass into advise(...) may be sent to whatever LLM provider you configure — the skill itself doesn't store or exfiltrate secrets, but your configured LLM/provider could. Also note the code parses LLM text with simple regexes, so malformed LLM responses can lead to fallback/default behavior (not a security issue but can affect accuracy).

功能分析

Type: OpenClaw Skill Name: decision-advisor Version: 1.0.0 The skill bundle implements a structured decision-making framework using the Tree of Thoughts methodology. The code in index.js focuses on generating options, defining criteria, and scoring alternatives via an LLM, with no evidence of data exfiltration, malicious execution, or prompt injection attempts.

能力评估

✓ Purpose & Capability

The name/description (Tree of Thoughts decision advisor) match the implementation: index.js implements option generation, criteria definition, scoring, and recommendation logic. There are no unexpected dependencies, environment variables, or external services declared.

✓ Instruction Scope

SKILL.md describes the decision workflow and shows JS usage that matches index.js. The runtime instructions and code only use the provided LLM interface and do not instruct reading system files, env vars, or posting to external endpoints.

✓ Install Mechanism

There is no install spec (instruction-only skill) and package.json is simple; nothing is downloaded or extracted. No high-risk install mechanism is present.

✓ Credentials

The skill declares no required environment variables or credentials. The only external interaction is via this.llm.generate — the host must supply an LLM implementation and any associated credentials; the skill itself does not request secrets.

✓ Persistence & Privilege

always is false and the skill does not modify other skills or system settings. It does not request permanent presence or elevated privileges.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install decision-advisor
安装完成后，直接呼叫该 Skill 的名称或使用 /decision-advisor 触发
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 advisor with Tree of Thoughts for multi-criteria analysis

元数据

Slug decision-advisor

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