AGENT SKILL
quant-orchestrator
Multi-Agent AI Quant System with factor mining, strategy generation, and automated backtesting
- License 许可
- MIT
- 最近更新
- 2026-06-10
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- quant-orchestrator
- metadata
- { "clawdbot": { "os": [ "darwin", "linux", "win32" ], "emoji": "📊", "install": [ { "run": "pip install lightgbm pandas numpy requests", "type": "script", "description": "Install dependencies" } ], "homepage": "https://clawhub.com/quant-orchestrator", "requires": { "env": [], "bins": [ "python", "pip" ] } } }
- description
- Multi-Agent AI Quant System with factor mining, strategy generation, and automated backtesting
03 SKILL.md 正文
Quant Orchestrator Skill
📊 Description
Multi-Agent AI Quant System for automated strategy research and backtesting.
💰 Pricing
- 1 USDC per call
- 1 USDT = 1000 tokens
- Min deposit: 8 USDT
- Platform fee: 5%
🚀 Features
- AI Factor Mining - Generate trading factors automatically
- AI Strategy Generation - Create strategies from factors
- Automated Backtesting - One-click backtest with metrics
- Strategy Evaluation - Sharpe, MaxDrawdown, WinRate, IC, IR
Architecture
Agentic Quant Platform
├── DataAgent (market data)
├── FactorAgent (AI factor mining)
├── StrategyAgent (AI strategy generation)
├── BacktestAgent (automated backtesting)
└── EvaluationAgent (strategy evaluation)
Usage
from skill_with_billing import QuantOrchestrator
# Run with user billing
orchestrator = QuantOrchestrator()
result = orchestrator.run_pipeline(
task="研究BTC动量策略",
user_id="USER_ID"
)
Output
- Sharpe Ratio
- Max Drawdown
- Win Rate
- IC (Information Coefficient)
- IR (Information Ratio)
- Strategy code
- Backtest results
Quick Start
# Install dependencies
pip install lightgbm pandas numpy requests
# Run basic pipeline
python skill.py "研究BTC动量策略"
# Run with billing
python skill_with_billing.py "研究BTC策略" user123
Files
skill.py- Basic skillskill_with_billing.py- Skill with paymentbilling.py- Payment SDKbtc_predictor_optimized.py- BTC prediction
04 同仓库更多 Skills
youtube-content
YouTube transcripts to summaries, threads, blogs.
★ 3
xurl
X/Twitter via xurl CLI: post, search, DM, media, v2 API.
★ 3
touchdesigner-mcp
Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.
★ 3
test-driven-development
TDD: enforce RED-GREEN-REFACTOR, tests before code.
★ 3
teams-meeting-pipeline
Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.
★ 3
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
★ 3