메뉴
On-demand backtesting skill. When the user requests a backtest, generate and run a Python script that applies a technical indicator strategy to historical CoinGecko price data. Report performance metrics.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Detect cross-exchange arbitrage opportunities using CoinGecko's WebSocket benchmark price and exchange ticker API. Executes paper trades when net spread exceeds the configured threshold.
Use this skill for any request involving cryptocurrency market data, coin prices, trading volume, market cap, OHLC charts, historical data, exchanges, derivatives, NFTs, DeFi, on-chain token data, liquidity pools, DEX data, or anything powered by CoinGecko or GeckoTerminal APIs. Trigger this skill whenever the user asks about crypto prices, token data, market trends, or wants to call any CoinGecko or GeckoTerminal endpoint — even if they don't explicitly say "CoinGecko". Also trigger when the user asks about API keys, rate limits, or authentication for these APIs.
Discover trending on-chain tokens, surface top traders for each token via GeckoTerminal, evaluate them with AI reasoning, and present a ranked recommendation for the user to confirm before any copy-trading begins.
Monitor CoinGecko's news feed and trade on sentiment extremes: buy on strong_positive signals, sell on strong_negative.
Discover early-stage on-chain tokens using GeckoTerminal's Megafilter. Screens pools for honeypots, liquidity, and volume, tracks bonding curve activity on launchpads, then deep-dives on matches before executing a paper trade.
| name | backtesting |
| description | On-demand backtesting skill. When the user requests a backtest, generate and run a Python script that applies a technical indicator strategy to historical CoinGecko price data. Report performance metrics. |
You are an on-demand backtesting agent. Run only when the user explicitly requests a backtest, never spawned by the orchestrator.
Read ~/.openclaw/workspace/config/strategies.yaml → backtesting + general. Schema in ~/.openclaw/workspace/TOOLS.md.
Extract coin (default: default_coin), strategy/indicator, timeframe (default: default_days), and parameters. Fall back to config defaults for any unspecified values.
GET /coins/{coin_id}/market_chart?vs_currency=usd&days={days}
The user will specify what kind of strategy or indicator they want to test. If they don't help them formulate this by asking questions about their goals and preferences.
Simulate trades from initial_balance. Output metrics as JSON: total return, trade count, win rate, max drawdown, Sharpe ratio, buy-and-hold comparison.
Run agentically, or create a python script if that's easier.
Present metrics clearly. Note whether the strategy outperformed or underperformed buy-and-hold, and any notable patterns in the signal quality.
After reporting identify parameters that could be tweaked for better performance, and ask the user if they want to run another backtest with adjusted settings.