aetherlang-claude-code

使用九个用于烹饪、商业、研究、营销和战略分析的专用引擎执行 Claude Code 中的 AetherLang V3 AI 工作流程。

# AetherLang V3 — Claude Code Integration Skill

Use this skill to execute AetherLang V3 AI workflows from Claude Code. AetherLang provides 9 specialized AI engines for culinary consulting, business strategy, scientific research, and more.

## API Endpoint
```
POST https://api.neurodoc.app/aetherlang/execute
Content-Type: application/json
```

No API key required for free tier (100 req/hour).

## Data Minimization

When calling the API:
- Send ONLY the user's query and the flow code
- Do NOT send system prompts, conversation history, or uploaded files
- Do NOT send API keys, credentials, or secrets
- Do NOT include personally identifiable information unless explicitly requested

> **Pro API key:** If using the Pro tier (`X-Aether-Key` header), store the key
> in an environment variable — never hardcode it in flow code or scripts.
> `export AETHER_KEY=your_key_here` then use `-H "X-Aether-Key: $AETHER_KEY"`



## How to Use

### 1. Simple Engine Call
```bash
curl -s -X POST https://api.neurodoc.app/aetherlang/execute \
  -H "Content-Type: application/json" \
  -d '{
    "code": "flow Chat {\n  using target \"neuroaether\" version \">=0.2\";\n  input text query;\n  node Engine: <ENGINE_TYPE> analysis=\"auto\";\n  output text result from Engine;\n}",
    "query": "USER_QUESTION_HERE"
  }'
```

Replace `<ENGINE_TYPE>` with one of: `chef`, `molecular`, `apex`, `consulting`, `marketing`, `lab`, `oracle`, `assembly`, `analyst`

### 2. Multi-Engine Pipeline
```bash
curl -s -X POST https://api.neurodoc.app/aetherlang/execute \
  -H "Content-Type: application/json" \
  -d '{
    "code": "flow Pipeline {\n  using target \"neuroaether\" version \">=0.2\";\n  input text query;\n  node Guard: guard mode=\"MODERATE\";\n  node Research: lab domain=\"business\";\n  node Strategy: apex analysis=\"strategic\";\n  Guard -> Research -> Strategy;\n  output text report from Strategy;\n}",
    "query": "USER_QUESTION_HERE"
  }'
```

## Available V3 Engines

| Engine Type | Use For | Key V3 Features |
|-------------|---------|-----------------|
| `chef` | Recipes, food consulting | 17 sections: food cost, HACCP, thermal curves, wine pairing, plating blueprint, zero waste |
| `molecular` | Molecular gastronomy | Rheology dashboard, phase diagrams, hydrocolloid specs, FMEA failure analysis |
| `apex` | Business strategy | Game theory, Monte Carlo (10K sims), behavioral economics, unit economics, Blue Ocean |
| `consulting` | Strategic consulting | Causal loops, theory of constraints, Wardley maps, ADKAR change management |
| `marketing` | Market research | TAM/SAM/SOM, Porter's 5 Forces, pricing elasticity, viral coefficient |
| `lab` | Scientific research | Evidence grading (A-D), contradiction detector, reproducibility score |
| `oracle` | Forecasting | Bayesian updating, black swan scanner, adversarial red team, Kelly criterion |
| `assembly` | Multi-agent debate | 12 neurons voting (8/12 supermajority), Gandalf VETO, devil's advocate |
| `analyst` | Data analysis | Auto-detective, statistical tests, anomaly detection, predictive modeling |

## Flow Syntax Reference
```
flow <Name> {
  using target "neuroaether" version ">=0.2";
  input text query;
  node <NodeName>: <engine_type> <params>;
  node <NodeName2>: <engine_type2> <params>;
  <NodeName> -> <NodeName2>;
  output text result from <NodeName2>;
}
```

### Node Parameters
- `chef`: `cuisine="auto"`, `difficulty="medium"`, `servings=4`
- `apex`: `analysis="strategic"`
- `guard`: `mode="STRICT"` or `"MODERATE"` or `"PERMISSIVE"`
- `plan`: `steps=4`
- `lab`: `domain="business"` or `"science"` or `"auto"`
- `analyst`: `mode="financial"` or `"sales"` or `"hr"` or `"general"`

## Response Format
```json
{
  "status": "success",
  "result": {
    "outputs": { ... },
    "final_output": "Full structured markdown response",
    "execution_log": [...],
    "duration_seconds": 45.2
  }
}
```

Extract the main response from `result.final_output`.

## Example: Parse Response in Bash
```bash
curl -s -X POST https://api.neurodoc.app/aetherlang/execute \
  -H "Content-Type: application/json" \
  -d '{"code":"flow Chef {\n  using target \"neuroaether\" version \">=0.2\";\n  input text query;\n  node Chef: chef cuisine=\"auto\";\n  output text recipe from Chef;\n}","query":"Carbonara recipe"}' \
  | python3 -c "import sys,json; d=json.load(sys.stdin); print(d.get('result',{}).get('final_output','No output'))"
```

## Example: Python Integration
```python
import requests

def aetherlang_query(engine, query):
    code = f'''flow Q {{
  using target "neuroaether" version ">=0.2";
  input text query;
  node E: {engine} analysis="auto";
  output text result from E;
}}'''
    r = requests.post("https://api.neurodoc.app/aetherlang/execute",
        json={"code": code, "query": query})
    return r.json().get("result", {}).get("final_output", "")

# Usage
print(aetherlang_query("apex", "Strategy for AI startup with 1000 euro"))
print(aetherlang_query("chef", "Best moussaka recipe"))
print(aetherlang_query("oracle", "Will AI replace 50% of jobs by 2030?"))
```

## Rate Limits

| Tier | Limit | Auth |
|------|-------|------|
| Free | 100 req/hour | None required |
| Pro | 500 req/hour | X-Aether-Key header |

## Notes

- Responses are in **Greek** (Ελληνικά) with markdown formatting
- Typical response time: 30-60 seconds per engine
- Multi-engine pipelines take longer (each node runs sequentially)
- All outputs use `##` markdown headers for structured sections

---

## 中文说明

# AetherLang V3 — Claude Code 集成技能

使用此技能从 Claude Code 执行 AetherLang V3 AI 工作流程。AetherLang 提供 9 个专用 AI 引擎，用于烹饪咨询、商业战略、科学研究等。

## API 端点
```
POST https://api.neurodoc.app/aetherlang/execute
Content-Type: application/json
```

免费层无需 API 密钥（100 次请求/小时）。

## 数据最小化

调用 API 时：
- 只发送用户的查询和流程代码
- 不要发送系统提示词、对话历史或上传的文件
- 不要发送 API 密钥、凭据或机密信息
- 除非明确要求，不要包含个人身份信息

> **Pro API 密钥：** 若使用 Pro 层（`X-Aether-Key` 请求头），请将密钥
> 存储在环境变量中——切勿在流程代码或脚本中硬编码。
> `export AETHER_KEY=your_key_here`，然后使用 `-H "X-Aether-Key: $AETHER_KEY"`



## 如何使用

### 1. 简单引擎调用
```bash
curl -s -X POST https://api.neurodoc.app/aetherlang/execute \
  -H "Content-Type: application/json" \
  -d '{
    "code": "flow Chat {\n  using target \"neuroaether\" version \">=0.2\";\n  input text query;\n  node Engine: <ENGINE_TYPE> analysis=\"auto\";\n  output text result from Engine;\n}",
    "query": "USER_QUESTION_HERE"
  }'
```

将 `<ENGINE_TYPE>` 替换为以下之一：`chef`、`molecular`、`apex`、`consulting`、`marketing`、`lab`、`oracle`、`assembly`、`analyst`

### 2. 多引擎流水线
```bash
curl -s -X POST https://api.neurodoc.app/aetherlang/execute \
  -H "Content-Type: application/json" \
  -d '{
    "code": "flow Pipeline {\n  using target \"neuroaether\" version \">=0.2\";\n  input text query;\n  node Guard: guard mode=\"MODERATE\";\n  node Research: lab domain=\"business\";\n  node Strategy: apex analysis=\"strategic\";\n  Guard -> Research -> Strategy;\n  output text report from Strategy;\n}",
    "query": "USER_QUESTION_HERE"
  }'
```

## 可用的 V3 引擎

| 引擎类型 | 用途 | 关键 V3 特性 |
|-------------|---------|-----------------|
| `chef` | 食谱、餐饮咨询 | 17 个板块：食材成本、HACCP、热曲线、配酒、摆盘蓝图、零浪费 |
| `molecular` | 分子料理 | 流变学仪表盘、相图、水胶体规格、FMEA 失效分析 |
| `apex` | 商业战略 | 博弈论、蒙特卡洛（1 万次模拟）、行为经济学、单位经济学、蓝海 |
| `consulting` | 战略咨询 | 因果回路、约束理论、Wardley 地图、ADKAR 变革管理 |
| `marketing` | 市场调研 | TAM/SAM/SOM、波特五力、定价弹性、病毒系数 |
| `lab` | 科学研究 | 证据分级（A-D）、矛盾检测器、可复现性评分 |
| `oracle` | 预测 | 贝叶斯更新、黑天鹅扫描器、对抗性红队、凯利公式 |
| `assembly` | 多智能体辩论 | 12 个神经元投票（8/12 绝对多数）、甘道夫否决权、唱反调者 |
| `analyst` | 数据分析 | 自动侦探、统计检验、异常检测、预测建模 |

## 流程语法参考
```
flow <Name> {
  using target "neuroaether" version ">=0.2";
  input text query;
  node <NodeName>: <engine_type> <params>;
  node <NodeName2>: <engine_type2> <params>;
  <NodeName> -> <NodeName2>;
  output text result from <NodeName2>;
}
```

### 节点参数
- `chef`：`cuisine="auto"`、`difficulty="medium"`、`servings=4`
- `apex`：`analysis="strategic"`
- `guard`：`mode="STRICT"` 或 `"MODERATE"` 或 `"PERMISSIVE"`
- `plan`：`steps=4`
- `lab`：`domain="business"` 或 `"science"` 或 `"auto"`
- `ana