메뉴
Guides development with TauricResearch TradingAgents multi-agent LLM financial trading framework (LangGraph, analyst teams, risk debate). Use for English multi-agent stock research, TradingAgentsGraph, propagate decisions, or porting agent roles to ION DEX AI analysis.
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
| name | trading-agents |
| description | Guides development with TauricResearch TradingAgents multi-agent LLM financial trading framework (LangGraph, analyst teams, risk debate). Use for English multi-agent stock research, TradingAgentsGraph, propagate decisions, or porting agent roles to ION DEX AI analysis. |
仓库: https://github.com/TauricResearch/TradingAgents
定位: LangGraph 多智能体 LLM 金融研究框架（分析师 → 研究员辩论 → 交易员 → 风控 → 组合经理）。
trading-agents-cn。TradingAgentsGraph + DEFAULT_CONFIG 驱动全流程。git clone https://github.com/TauricResearch/TradingAgents.git
cd TradingAgents
pip install -e .
# 配置 OPENAI_API_KEY 等，见仓库 .env.example
from tradingagents.graph.trading_graph import TradingAgentsGraph
from tradingagents.default_config import DEFAULT_CONFIG
ta = TradingAgentsGraph(debug=True, config=DEFAULT_CONFIG.copy())
_, decision = ta.propagate("AAPL", "2026-05-27")
print(decision)
AgentState 字段一致。config 中 provider/model；注意 API 调用次数高。propagate 输出映射为 UI 可展示的「观点 + 置信度 + 风险标签」，不要直接下单。详见 references.md。
Public stub — full GitHub daily discovery Skill lives in ion-private-core (CONFIDENTIAL). Use for 全品类开发流程 GitHub 高星发现（AI/MCP/Web3/前端/测试/视频/量化/CI/CD/数据库/K8s 等），每类 Top5 榜、DEX curated 全量存根、vendor 浅克隆。
Guides 51bitquant ai-hedge-fund-crypto LangGraph DAG workflow with multi-timeframe technical ensemble and LLM portfolio decisions for cryptocurrency. Use for crypto-specific agent graphs, strategy ensembling, or ION DEX AI trading research modules.
Enforces Chinese as the default language for ION DEX planning, development, testing, audit reports, delivery summaries, and user communication. Use for all ION DEX work unless the user explicitly requests another language.
Uses Claude-Flow/RuFlo 3.7.0-alpha.35 as a controlled AI agent orchestration accelerator for planning, multi-agent decomposition, swarm experiments, MCP diagnostics, memory workflows, and code analysis. Use when the user mentions Claude-Flow, RuFlo, 98 AI agents, swarm, hive-mind, autopilot, or agent orchestration.
Applies Cursor official-documentation workflows to ION DEX development. Use proactively for every development, verification, review, debugging, workflow, or tooling task to choose the right Cursor mode, Agent Review/Bugbot path, Rules/Skills/MCP/Hooks strategy, Cloud Agent/CLI/SDK automation, safety controls, and verification plan.
Finds, lists, and inspects available Cursor Agent Skills in project and user skill directories. Use when the user asks what skills exist, wants a skill installed, wants to locate a workflow, or when the agent needs to choose the right skill before acting.