MCP 助手

AI 智能社区

作者仓库星标 2,980
作者仓库 tradingview-mcp

解读按原文结构重写，命令、链接、术语均保留；右侧可核对作者原始 SKILL.md

原文结构

How to Run Trading Tools：Execute via bash using the wrapper script:
Behavior Guidelines：Run bash immediately. For any trading/market question → execute the command directly, don't ask for clarification. Always combine signals. For "should I buy X?" → run price + backtest + sentiment together. Qualify with timeframe. Default to 1y period and 1d…
Tool Quick Reference：原文以此为独立章节。

适用与边界

当前原文没有单列适用场景。
当前原文没有单列不适用场景或限制。

原文中的明确线索

要点：「NEVER use sessionsspawn or ask for an agent ID for trading tasks.」、「Run bash immediately.」、「Always combine signals.」、「Qualify with timeframe.」、「Explain metrics briefly.」、「Add a disclaimer」、「Be concise on Telegram.」、「Detect language.」
文件与命令：sessionsspawn、yahooprice(symbol="AAPL")、getpricesbulk(symbols=["BTC-USD","ETH-USD"])、marketsnapshot()、technicalanalysis(symbol="AAPL", exchange="NASDAQ", screener="america", interval="1h")、calculatersi(symbol="BTC-USD", period="14")、calculatesupertrend(symbol="AAPL")、backteststrategy(symbol="AAPL", strategy="rsi", period="1y")

流狐整理：以上内容来自当前 SKILL.md 的章节与原词；未补写作者没有声明的工具、兼容性或能力。

TradingView MCP — AI Trading Intelligence

You have a trading intelligence tool available via bash. NEVER use sessions_spawn or ask for an agent ID for trading tasks. Run commands directly.

Use this tool whenever users ask about:

Stock, crypto, ETF, or index prices
Technical analysis (RSI, MACD, Bollinger Bands, etc.)
Backtesting trading strategies
Market sentiment or news
Screening for trading opportunities

How to Run Trading Tools

Execute via bash using the wrapper script:

python3 ~/.openclaw/tools/trading.py <command> [args]

Behavior Guidelines

Run bash immediately. For any trading/market question → execute the command directly, don't ask for clarification.
Always combine signals. For "should I buy X?" → run price + backtest + sentiment together.
Qualify with timeframe. Default to 1y period and 1d interval unless specified.
Explain metrics briefly. Sharpe (risk-adjusted return), Max Drawdown (worst loss), Profit Factor (wins/losses).
Add a disclaimer on all backtests: "⚠️ Past performance does not guarantee future results."
Be concise on Telegram. Use emoji, bullet lists — no walls of JSON.
Detect language. Reply in the same language the user writes in.

Tool Quick Reference

Prices & Market

Intent	Tool
"What is AAPL's price?"	`yahoo_price(symbol="AAPL")`
"Show me BTC and ETH prices"	`get_prices_bulk(symbols=["BTC-USD","ETH-USD"])`
"How are markets today?"	`market_snapshot()`

Technical Analysis

Intent	Tool
"Analyze AAPL technically"	`technical_analysis(symbol="AAPL", exchange="NASDAQ", screener="america", interval="1h")`
"What is the RSI for BTC?"	`calculate_rsi(symbol="BTC-USD", period="14")`
"Supertrend signal for AAPL?"	`calculate_supertrend(symbol="AAPL")`

Backtesting

Intent	Tool
"Backtest RSI strategy for 1 year"	`backtest_strategy(symbol="AAPL", strategy="rsi", period="1y")`
"Show me the full trade log"	`backtest_strategy(symbol="BTC-USD", strategy="supertrend", period="1y", include_trade_log=True)`
"Run hourly backtest"	`backtest_strategy(symbol="AAPL", strategy="bollinger", period="3mo", interval="1h")`
"Which strategy is best?"	`compare_strategies(symbol="BTC-USD", period="2y")`
"Is this strategy overfitted?"	`walk_forward_backtest_strategy(symbol="AAPL", strategy="rsi", period="2y", n_splits=3)`

Sentiment & News

Intent	Tool
"What is Reddit saying about BTC?"	`analyze_sentiment(symbol="BTC")`
"Latest news on AAPL"	`fetch_news_summary(symbol="AAPL")`
"Combine technical + sentiment"	`analyze_confluence(symbol="AAPL", exchange="NASDAQ")`

Screener

Intent	Tool
"Strong bullish stocks"	`screener_bullish(exchange="NASDAQ")`
"Find oversold stocks"	`screener_oversold(exchange="NASDAQ")`
"Scan Turkish BIST stocks"	`screener_bullish(exchange="BIST")`
"Egyptian Exchange stocks"	`egx_stock_screen()`

Example Response Formats

Price Query (Telegram-friendly)

📊 AAPL — Apple Inc.
💵 Price: $189.42
📈 Change: +1.23% (+$2.30)
📅 52w High: $199.62 | Low: $164.08
🏦 Exchange: NASDAQ | Market: REGULAR

Backtest Summary (Telegram-friendly)

🔬 AAPL — RSI Strategy (1Y daily)
────────────────────────────────
📊 Trades: 8 | Win Rate: 62.5%
💰 Return: +14.3% vs B&H: +21.2%
📉 Max Drawdown: -6.8%
⚡ Sharpe: 1.42 | Calmar: 2.10
🏆 Profit Factor: 2.31

⚠️ Past performance does not guarantee future results.

Walk-Forward (Overfitting Check)

🧪 Walk-Forward: AAPL RSI (2Y, 3 folds)
────────────────────────────────────────
Robustness Score: 0.87 → ROBUST ✅
Train avg: +12.4% | Test avg: +10.8%

Fold 1: Train +18% → Test +15% (rob: 0.83)
Fold 2: Train +8%  → Test +7%  (rob: 0.88)
Fold 3: Train +11% → Test +10% (rob: 0.91)

✅ Strategy performs well out-of-sample.

Supported Symbols

US Stocks: AAPL, TSLA, NVDA, MSFT, GOOGL, META, AMZN
Crypto: BTC-USD, ETH-USD, SOL-USD, BNB-USD, XRP-USD
ETFs: SPY, QQQ, GLD, VTI, IWM
Indices: ^GSPC (S&P500), ^IXIC (NASDAQ), ^DJI (Dow), ^VIX
Turkish (BIST): THYAO.IS, SASA.IS, BIMAS.IS, KCHOL.IS, EKGYO.IS
Egyptian (EGX): COMI.CA, HRHO.CA, EAST.CA
FX: EURUSD=X, GBPUSD=X, JPYUSD=X, TRYUSD=X

Strategies Available for Backtesting

Strategy	Key	Best For
RSI Mean Reversion	`rsi`	Ranging/sideways markets
Bollinger Band	`bollinger`	Mean reversion in volatile markets
MACD Crossover	`macd`	Trend following
EMA 20/50 Cross	`ema_cross`	Medium-term trends
Supertrend (ATR)	`supertrend`	Strong trending markets
Donchian Channel	`donchian`	Breakout / Turtle Trading

流狐档案作者与许可取自来源；运行、权限和网络为流狐检测或估算

流狐分类: AI 智能
作者声明 Agent: 未找到明确声明；不据此推断已兼容或已测试
静态检查: 88 / 100 · 启发式扫描，不代表运行安全
作者 / 版本 / 许可: @atilaahmettaner · 未声明 license
流狐 Token 估算: 低消耗
流狐接入估算: 需简单配置
是否需要外部 API Key: 未发现要求
检测到的系统要求: 未声明
底层运行要求: 未声明
检测到的文件与系统行为: 只读
允许写入 / 修改
Shell 执行
检测到的网络行为: 仅限本地
安装命令数: 无（仅作为资料）

档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。

需要注意： 未限定 allowed-tools，默认拥有全部工具权限。

输出预览 tradingview-mcp.preview

作者没有在当前 SKILL.md 中定义固定输出样例。

SKILL.md SKILL.md

# TradingView MCP — AI Trading Intelligence

You have a trading intelligence tool available via bash. **NEVER use `sessions_spawn` or ask for an agent ID for trading tasks.** Run commands directly.

Use this tool whenever users ask about:
- Stock, crypto, ETF, or index prices
- Technical analysis (RSI, MACD, Bollinger Bands, etc.)
- Backtesting trading strategies
- Market sentiment or news
- Screening for trading opportunities

## How to Run Trading Tools

Execute via bash using the wrapper script:
```bash
python3 ~/.openclaw/tools/trading.py <command> [args]
```

## Behavior Guidelines

1. **Run bash immediately.** For any trading/market question → execute the command directly, don't ask for clarification.
2. **Always combine signals.** For "should I buy X?" → run price + backtest + sentiment together.
3. **Qualify with timeframe.** Default to `1y` period and `1d` interval unless specified.
4. **Explain metrics briefly.** Sharpe (risk-adjusted return), Max Drawdown (worst loss), Profit Factor (wins/losses).
5. **Add a disclaimer** on all backtests: "⚠️ Past performance does not guarantee future results."
6. **Be concise on Telegram.** Use emoji, bullet lists — no walls of JSON.
7. **Detect language.** Reply in the same language the user writes in.

## Tool Quick Reference

### Prices & Market
| Intent | Tool |
|--------|------|
| "What is AAPL's price?" | `yahoo_price(symbol="AAPL")` |
| "Show me BTC and ETH prices" | `get_prices_bulk(symbols=["BTC-USD","ETH-USD"])` |
| "How are markets today?" | `market_snapshot()` |

### Technical Analysis
| Intent | Tool |
|--------|------|
| "Analyze AAPL technically" | `technical_analysis(symbol="AAPL", exchange="NASDAQ", screener="america", interval="1h")` |

…

证据边界与执行链路

作者原文负责流程事实；流狐只索引当前章节、要点、文件与命令。

章节    -> How to Run Trading Tools → Behavior Guidelines → Tool Quick Reference → Prices & Market → Technical Analysis → Backtesting
要点    -> NEVER use sessionsspawn or ask for an agent ID for trading tasks. · Run bash immediately. · Always combine signals. · Qualify with timeframe. · Explain metrics briefly. · Add a disclaimer · Be concise on Telegram. · Detect language.
文件/命令 -> sessionsspawn · yahooprice(symbol="AAPL") · getpricesbulk(symbols=["BTC-USD","ETH-USD"]) · marketsnapshot() · technicalanalysis(symbol="AAPL", exchange="NASDAQ", screener="america", interval="1h") · calculatersi(symbol="BTC-USD", period="14") · calculatesupertrend(symbol="AAPL") · backteststrategy(symbol="AAPL", strategy="rsi", period="1y")
内容 SHA-256 -> 1accf4e1e2a4