A sophisticated multi-agent AI framework that orchestrates 15 specialized analysts to deliver institutional-grade research and trading recommendations for China A-shares.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install tradingagents-analysis
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install tradingagents-analysis using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is TradingAgents Multi-Agent A-Share Analysis?
TradingAgents is a professional-grade investment research tool specifically designed for the China A-share market. By utilizing Openclaw Skills, it coordinates a swarm of 15 specialized AI agents to conduct comprehensive, five-stage collaborative research. This goes far beyond basic price alerts, offering a holistic view that includes technical analysis, fundamental data, market sentiment, and macroeconomic trends.
Built to handle the unique complexities of the Shanghai and Shenzhen stock exchanges, this skill leverages the power of Openclaw Skills to automate deep-dive reports. It simulates an institutional research desk by involving different personas—such as bull and bear researchers, risk managers, and game theory specialists—to ensure that every recommendation is pressure-tested against opposing viewpoints and market risks.
TradingAgents Multi-Agent A-Share Analysis Use Cases
- Perform deep-dive technical and fundamental analysis on specific A-share stocks.
- Generate structured buy, sell, or hold recommendations backed by multi-agent consensus.
- Conduct post-market reviews and risk assessments for existing portfolios.
- Track smart money flow, including Northbound capital and main-force movements.
- Analyze market sentiment and retail-vs-institutional expectation gaps using game theory.
How TradingAgents Multi-Agent A-Share Analysis Works
- The user identifies a stock by name or symbol and initiates the request through Openclaw Skills.
- The system triggers the Analyst Team to gather raw data on macro trends, fundamentals, and news.
- A Game Theory manager analyzes the expectation gap between main-force capital and retail investors.
- Specialized Bull and Bear agents engage in an adversarial debate to challenge the prevailing thesis.
- A Trade Execution agent synthesizes the debate and data into a concrete action plan.
- Multiple Risk Control agents (Aggressive, Neutral, and Conservative) audit the decision for safety.
- The final structured JSON report is delivered to the user with target prices and stop-loss levels.
TradingAgents Multi-Agent A-Share Analysis Setup
To integrate this capability with Openclaw Skills, follow these installation steps:
- Generate an API token at app.510168.xyz.
- Configure your environment variables:
export TRADINGAGENTS_TOKEN="ta-sk-your_key_here"
# Optional: Point to a self-hosted backend
export TRADINGAGENTS_API_URL="https://api.510168.xyz"
- Execute the analysis script provided in the repository:
bash scripts/analyze.sh <stock_symbol>
TradingAgents Multi-Agent A-Share Analysis Data Schema & Taxonomy
The skill returns a highly structured data object ensuring clarity for Openclaw Skills users:
| Key | Type | Description |
|---|---|---|
decision |
String | The final recommendation (e.g., BUY, SELL, HOLD). |
confidence |
Integer | A percentage score (0-100) representing agent consensus. |
target_price |
Float | The projected price target for the analysis horizon. |
risk_items |
Array | A list of identified risks with severity levels (Low/Medium/High). |
key_metrics |
Array | Financial data points like PE, ROE, and Gross Margin with status indicators. |
final_trade_decision |
String | A textual summary of the synthesized multi-agent logic. |
TradingAgents Multi-Agent A-Share Analysis Advanced Features
- Multi-stock batch processing allowing for parallel analysis of entire watchlists.
- Support for self-hosted backend deployments to maintain total data sovereignty within Openclaw Skills.
- Short-term trading mode that prioritizes 14-day technical indicators and skips heavy financial reporting for speed.
- Configurable polling logic and timeout settings to handle deep-research tasks that take 1-5 minutes.
- Integrated adversarial modeling where Bull and Bear agents debate to eliminate confirmation bias.
SKILL.md
Loading
Related Openclaw Skills
Groupon
An AI-driven assistant for finding, vetting, and managing Groupon vouchers through rigorous fine-print analysis and merchant risk assessment.
Plex Media Server
A comprehensive toolset for managing Plex libraries and sessions through Openclaw Skills.
Ecommerce Image Asset Generator
A strategic AI tool for planning and generating high-conversion ecommerce visual assets that align with specific business objectives.
Critic Agent Skill
A specialized agent for reviewing, scoring, and providing actionable feedback on AI-generated outputs using a structured rubric.
Git Toolkit
A comprehensive set of extended Git utilities for repository analysis, automation, and advanced version control management.
Synology
A professional framework for planning, hardening, and recovering Synology NAS environments with safety-first DSM workflows.
Featured*
