Agent Skills › winstonkoh87/Athena-Public › Decision Matrix (MCDA)

Decision Matrix (MCDA)

GitHub

多准则决策分析计算器，将复杂决策拆解为加权标准与明确评分。适用于多选项权衡、利益相关者分歧场景，通过计算期望价值(EEV)量化推荐方案，并执行敏感性分析以验证结果稳健性。

examples/skills/decision/mcda-solver/SKILL.md winstonkoh87/Athena-Public

Trigger Scenarios

需要在多个选项中基于多重标准进行权衡决策利益相关者对优先级存在分歧或难以达成共识面临复杂的‘视情况而定’型选择问题

Install

npx skills add winstonkoh87/Athena-Public --skill Decision Matrix (MCDA) -g -y

More Options

Non-standard path

npx skills add https://github.com/winstonkoh87/Athena-Public/tree/main/examples/skills/decision/mcda-solver -g -y

Use without installing

npx skills use winstonkoh87/Athena-Public@Decision Matrix (MCDA)

指定 Agent (Claude Code)

npx skills add winstonkoh87/Athena-Public --skill Decision Matrix (MCDA) -a claude-code -g -y

安装 repo 全部 skill

npx skills add winstonkoh87/Athena-Public --all -g -y

预览 repo 内 skill

npx skills add winstonkoh87/Athena-Public --list

SKILL.md

Frontmatter

{
    "name": "Decision Matrix (MCDA)",
    "model": "default",
    "created": 1772150400,
    "auto-invoke": false,
    "description": "Multi-Criteria Decision Analysis calculator. Breaks complex decisions into weighted criteria with explicit scoring.",
    "context_trigger": "decision matrix, multi-criteria, MCDA, weighted scoring, trade-off analysis, alternatives, prioritization"
}

⚖️ Decision Matrix (MCDA / EEV)

Philosophy: Complex decisions fail because people optimize for one variable while ignoring five others.

1. When to Use

Choosing between 2+ options with multiple trade-offs
Any decision where "it depends" is the current answer
When stakeholders disagree on priorities

2. Execution Workflow

STEP 1: LIST OPTIONS
  └─ Must have ≥2 distinct options (no straw men)

STEP 2: DEFINE CRITERIA
  └─ List 3-7 evaluation criteria
  └─ Each criterion must be measurable or scorable (1-10)

STEP 3: WEIGHT CRITERIA
  └─ Assign percentage weights (must sum to 100%)
  └─ Force-rank: "If you could only optimize for ONE criterion, which?"

STEP 4: SCORE EACH OPTION
  └─ Score 1-10 per criterion per option
  └─ Justify each score in one sentence

STEP 5: COMPUTE EEV
  └─ EEV = Σ (Score × Weight) for each option
  └─ Rank options by total EEV

STEP 6: SANITY CHECK
  └─ Does the winner "feel" right?
  └─ If not, a hidden criterion exists — surface it and re-run

3. Output Format

# Decision: [Question]

## Criteria & Weights
| # | Criterion | Weight |
|---|-----------|--------|
| 1 | [Name]    | [X]%   |
| 2 | [Name]    | [X]%   |

## Scoring Matrix
| Option | C1 | C2 | C3 | **EEV** |
|--------|----|----|-----|---------|
| A      | 8  | 6  | 7   | **7.1** |
| B      | 5  | 9  | 4   | **6.2** |

## Recommendation
[Winner] with EEV of [X]. Key differentiator: [criterion].

## Sensitivity Analysis
If [criterion weight] changed by ±10%, would the winner change? [Yes/No]