SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillstradingagents-analysis

tradingagents-analysis

Agents & models

View repository

Updated 10-15 23:48

About this skill

TradingAgents Multi-Agent A-Share Investment Research Analysis

Using the TradingAgents API, let 15 professional AI analystsconduct five-stage in-depth collaborative research and judgment on A-shares, and output structured investment recommendations.

🎯 Quick Start

Just say to me:

  • "Help me analyze Kweichow Moutai"
  • "Is CATL worth buying"
  • "Analyze the technicals of 600519"
  • "How has BYD's capital flow been recently"

I will call 15 AI analysts to conduct in-depth analysis from five dimensions: market, technical, fundamental, sentiment, and capital, and give you professional investment recommendations.

🤖 System Architecture: Five Stages, 15 Agents

Stage Agent Responsibility
1. Analysis Team Market/News/Sentiment/Fundamental/Macro/Smart Money Multi-dimensional raw data interpretation
2. Game Referee Game Theory Manager Analysis of expectation gaps between main capital and retail investors
3. Bull-Bear Debate Bull/Bear Researcher + Referee Intense game of opposing views
4. Execution Decision Trader Comprehensive research and judgment to generate operational recommendations
5. Risk Control Aggressive/Neutral/Conservative Analyst + Portfolio Manager Multi-dimensional risk control review
TradingAgents Multi-Agent Investment Research

Use the TradingAgents API to let 15 specialized AI analysts conduct deep, five-stage collaborative research on A-Share stocks, delivering structured trading recommendations.

🤖 System Architecture: 5 Stages · 15 Agents

Stage Agents Role
1. Analyst Team Market / News / Sentiment / Fundamentals / Macro / Smart Money Multi-dimensional raw data analysis
2. Game Theory Game Theory Manager Main-force vs. retail expectation gap
3. Bull/Bear Debate Bull & Bear Researchers + Judge Adversarial viewpoint debate
4. Trade Execution Trader Synthesize research into actionable decision
5. Risk Control Aggressive / Neutral / Conservative + Portfolio Manager Multi-layer risk review

📋 Applicable Scenarios

✅ Suitable for use:

  • In-depth individual stock analysis (technical + fundamental)
  • Investment decision reference
  • After-hours review analysis
  • Risk assessment of held positions
  • Capital flow and market sentiment assessment

❌ Not suitable for:

  • Real-time intraday monitoring (analysis takes 1-5 minutes)
  • Ultra-short-term trading (minute-level decisions)
  • Non-A-share markets such as cryptocurrencies and US stocks

🔒 Privacy and Security

  • Sending Scope: This skillonlyextracts parameters such as stock name/code, analysis date, and analysis perspective from the conversation, and sends them as the symbol/trade_date/horizons field to the backend API.It does not send the original conversation text, does not read local files, and does not upload any other private data.
  • Token Security：TRADINGAGENTS_TOKEN(format ta-sk-*) is the only credential for accessing the backend. Please use a least-privilege token. If you suspect a leak, immediately revoke it at app.510168.xyz and regenerate it.
  • Sensitive Content Notice: Please do not paste personal account information, real holdings, or other sensitive content into analysis requests. This skill cannot prevent users from actively submitting such content.
  • Self-Hosting: If you need full control over data flow, you can refer to the GitHub documentation to deploy the backend yourself, and point TRADINGAGENTS_API_URL to your self-hosted server.

About credential metadata: The frontmatter of this skill is in metadata.openclaw declared TRADINGAGENTS_TOKEN as primaryEnv, and is listed in requires.env。

🔒 Privacy & Data Transmission

  • What is sent: Only the extracted stock symbol, trade date, and analysis parameters (symbol, trade_date, horizons) are transmitted to the backend. The raw conversation text is never forwarded.
  • Token: TRADINGAGENTS_TOKEN (pattern ta-sk-*) is the sole credential. Use a minimal-privilege token and rotate it immediately if compromised.
  • Sensitive content: Do not paste personal account data, real positions, or other sensitive information into analysis requests.
  • Self-hosting: For full data sovereignty, deploy the backend yourself and set TRADINGAGENTS_API_URL to your server. See the GitHub repo.

Credential metadata: This skill's frontmatter declares TRADINGAGENTS_TOKEN as primaryEnv under metadata.openclaw.requires.env.

⚙️ Quick Configuration

Option 1: Use the official hosted service (zero deployment, ready to use out of the box)

  1. Log in to https://app.510168.xyz
  2. and go to Settings → API Tokens to create a token
  3. Configure environment variables:
export TRADINGAGENTS_TOKEN="ta-sk-your_key_here"

Option 2: Private deployment (fully autonomous control of data)

If you have data privacy requirements, you can deploy the backend yourself; all analysis data will be processed only on your own server:

# 1. 部署后端，参考 https://github.com/KylinMountain/TradingAgents-AShare
# 2. 将 API 地址指向自建服务
export TRADINGAGENTS_API_URL="http://your-server:8000"
export TRADINGAGENTS_TOKEN="ta-sk-your_key_here"

🚀 Common Operations

Recommended approach: Use the all-in-one script(automatically submit → poll → get results)

# 脚本路径（相对于技能目录）
bash scripts/analyze.sh <symbol[,symbol2,...]> [trade_date] [horizons]

# 单个分析
bash scripts/analyze.sh 贵州茅台
bash scripts/analyze.sh 600519.SH 2026-03-22
bash scripts/analyze.sh 600519.SH 2026-03-22 medium

# 批量分析（逗号分隔，并行提交，统一等待）
bash scripts/analyze.sh 贵州茅台,比亚迪,宁德时代
bash scripts/analyze.sh 600519.SH,002594.SZ,300750.SZ 2026-03-22

The script automatically completes: submit task → poll status every 15 seconds → output JSON results when complete. In batch mode, all tasks are submitted in parallel, polled uniformly, and summarized in the final output. The default timeout is 600 seconds.

Behavior can be adjusted via environment variables:

  • POLL_INTERVAL — polling interval in seconds (default 15)
  • POLL_TIMEOUT — maximum waiting time in seconds (default 600)

Manual step-by-step operations(if you need to call a step separately)

All requests use $TRADINGAGENTS_TOKEN as the Bearer token.

  1. Submit analysis task
curl -X POST "${TRADINGAGENTS_API_URL:-https://api.510168.xyz}/v1/analyze" \
  -H "Authorization: Bearer $TRADINGAGENTS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"symbol": "贵州茅台"}'
  1. Query task status
curl "${TRADINGAGENTS_API_URL:-https://api.510168.xyz}/v1/jobs/{job_id}" \
  -H "Authorization: Bearer $TRADINGAGENTS_TOKEN"
  1. Get full analysis results (after task completion)
curl "${TRADINGAGENTS_API_URL:-https://api.510168.xyz}/v1/jobs/{job_id}/result" \
  -H "Authorization: Bearer $TRADINGAGENTS_TOKEN"

📊 Example Output

{
  "decision": "BUY",
  "direction": "看多",
  "confidence": 78,
  "target_price": 1850.0,
  "stop_loss_price": 1680.0,
  "risk_items": [
    {"name": "估值偏高", "level": "medium", "description": "当前 PE 处于历史 75 分位"},
    {"name": "外资流出", "level": "low",    "description": "近 5 日北向资金小幅净流出"}
  ],
  "key_metrics": [
    {"name": "PE",   "value": "32.5x",  "status": "neutral"},
    {"name": "ROE",  "value": "31.2%",  "status": "good"},
    {"name": "毛利率", "value": "91.5%", "status": "good"}
  ],
  "final_trade_decision": "综合技术面突破与基本面支撑，建议逢低分批建仓..."
}

🔄 Task Execution Flow

Deep analysis usually takes 1 to 5 minutes：

  1. Identify the target: From the conversation,onlyextract the stock name or ticker (and optionally date/perspective); do not send the original conversation text
  2. Inform the user: Give feedback that the task is about to be submitted and is expected to take 1-5 minutes
  3. Execute the script: Use the Bash tool to run bash scripts/analyze.sh <symbol> [date] [horizons](set run_in_background: true); the script automatically completes submission, polling, and result retrieval
  4. Summarize the conclusion: After the script output completes, parse the JSON results and show the user the decision, direction, target price, and risk points

Important: Do not manually write a curl polling loop; directly use the scripts/analyze.sh script.

📌 Supported Target Range

  • Shanghai and Shenzhen A-shares: Chinese names (such as "BYD", "CATL") or tickers (002594.SZ、601012.SH）

💡 Notes

  • Polling frequency: Each polling interval must be no less than 15 seconds
  • Data robustness: If some data sources are missing, the system will perform spillover analysis based on macro and industry logic
  • Short-term mode: When entering "analyze XX short-term", the system automatically switches to 14-day technical analysis, skips financial report data, and is faster