📊 Full opportunity report: Introducing Forezai · TradingAgents — a committee of LLMs decides paper-trades on ThorstenMeyerAI.com — validation score, market gap, and execution plan.

TL;DR

Forezai has introduced TradingAgents, a system where a committee of specialized large language models (LLMs) collaboratively generate paper-trading decisions. This development aims to explore whether AI teams can outperform random decision-making in simulated trading environments.

Forezai has launched TradingAgents, a software framework where a committee of large language models (LLMs) collaboratively evaluate market data and generate paper-trading decisions. This initiative aims to test whether AI-driven multi-agent systems can produce decision quality comparable to or better than random chance, marking a step forward in AI research for financial decision-making.

The project, a fork of the open-source TradingAgents framework developed by TauricResearch, adds operational features to enable autonomous, scheduled trading simulations. It includes a scheduler, an auto-trader that maps AI ratings to paper orders, and a multi-broker abstraction supporting local, Alpaca, and shadow modes. The system features a web dashboard for monitoring performance, all running locally without cloud data transmission.

Unlike single-model predictions, TradingAgents employs a structured multi-role approach: four analyst agents generate independent reports on market structure, news flow, fundamentals, and social sentiment; two debate opposing theses; a research manager synthesizes these reports; and a risk team evaluates upside and downside. The final decision-making layer produces a five-tier rating with target prices and time horizons. The design emphasizes explicit reasoning and argumentation over raw prediction.

Forezai’s development emphasizes research rather than live trading, with safeguards against real-money risks. The system’s architecture is intended for experimentation, enabling researchers to assess whether collective AI reasoning can surpass random or traditional rule-based strategies in simulated environments.

Introducing Forezai · TradingAgents — Thorsten Meyer AI
AGENTS
● ANNOUNCEMENT / MAY 2026
THORSTEN MEYER AI · FOREZAI · § 03
FOREZAI · 03
TRADINGAGENTS · LAUNCH
Research Series · Companion to Polybot Week 1-2 · 2026-05-17

Introducing Forezai · TradingAgents.
A committee of LLMs
decides paper-trades.

After two weeks of finding out most parametric strategies don’t work, the obvious next research question: can multi-agent LLM judgment do any better?
A fork of the open-source TradingAgents framework (TauricResearch): thirteen LLM agents in four stages — four parallel analysts · a bull-bear debate with research-manager arbitration · a three-voice risk team · a two-layer trader + portfolio-manager decision. The fork keeps the agent graph intact and adds the operational layer the upstream doesn’t ship: an autonomous loop · a multi-broker abstraction · a local web dashboard · Codex OAuth · MCP plug-ins · 520+ unit tests. The question is narrower than “do LLMs predict the market” — that prior is “no, with high confidence.” The narrower question is: when LLMs are structured into specialised adversarial roles, does the committee produce decisions at least no worse than a coin flip after fees? Honest priors before running: it might fail too. If it appears to work, the most likely explanation is variance.
This is not financial advice. Nothing in this announcement should be used to inform real trading decisions. The software described trades simulated money by default. If you reconfigure it to trade real money, you should expect to lose that money — regardless of how clever any individual agent’s reasoning looks. Algorithmic trading is zero-sum after fees and structurally hostile to part-time retail strategies.
13 agents
Specialised roles in four stages
Analysts · Debate · Risk · Decision
78% / -33%
Polybot prior: fleet win rate
combined with -33% bankroll
520+
Passing unit tests across engine,
services, HTTP routes (starting baseline)
€0 floor
LLM cost on Codex OAuth
(falls back to public API per token)
FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE· FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE·
FIG. 01 — THE 13-AGENT COMMITTEE
Thirteen specialised roles · four stages · biases made to argue in public
The architecture forces the system to articulate its reasoning rather than relying on what a single context window happens to recall
Stage 1 · Four analysts in parallel4 agents
Market
Structure, ranges, regime indicators
News + Insider
News flow, filings, insider activity
Fundamentals
Balance sheet, earnings, ratios
Social Sentiment
Social-media tone, retail signal
Stage 2 · Bull-bear debate + research-manager arbitration3 agents
Bull researcher
Argues upside thesis from analyst reports
Bear researcher
Argues downside thesis from same reports
Research manager
Arbitrates · writes single synthesis
Stage 3 · Three-voice risk team3 agents
Aggressive
Looks for upside · accepts variance
Conservative
Looks for downside · protects capital
Neutral
Balances · forces downside articulation
Stage 4 · Two-layer decision2 agents
Trader
Three-tier proposal · buy / hold / sell
Portfolio manager
Five-tier rating + price target + horizon · sees arguments only, never raw data
The portfolio manager only sees the arguments, never the raw data — which forces the committee to make its reasoning explicit rather than relying on a single context window’s recall. The upstream framework ships the agent graph; it does not ship the operational machinery to run that graph on autopilot, observe its results honestly, store them for later inspection, or prevent the operator from accidentally trading real money. That gap is what the Forezai fork fills.
FIG. 02 — THE POLYBOT PRIOR · WHY THIS IS A DIFFERENT BET
Two weeks of paper-trading prediction markets · the trap underneath the headline numbers
25 experiments · 78% fleet-wide win rate · -33% bankroll · most parametric strategies are structurally negative-expectation when measured honestly
The flattering number
78%
Fleet-wide win rate · week 2
“You can win four out of five trades and still go broke, because the one loss is bigger than the four wins put together.” Win rate without P&L context is a mechanical illusion.
The honest number
−33%
Fleet bankroll · week 2 close
The strongest possible demonstration of the trap. A parametric trading strategy that looks compelling in a backtest will almost always fail to survive a fresh sample. Most “edges” are mechanical artefacts.
Week 1: 21 parallel strategy experiments · early winners mostly mechanical illusions · exactly one strategy (a fair-value taker on BTC) showed the mathematical signature of real edge over a few hundred settled trades. Week 2: same fair-value strategy with more data collapsed. A separate mid-week hypothesis (market-making) also failed cleanly. Fleet ended week 2 at roughly negative thirty-three percent of bankroll. The honest research finding wasn’t on the winning side — it was on the losing side. Adding more parameters to Polybot wouldn’t change that. TradingAgents is asking a separable question.
FIG. 03 — WHAT THE FORK ADDS · THE OPERATIONAL LAYER
Six layers the upstream framework doesn’t ship
Same agent graph, intact. The fork makes it a research instrument rather than a tech demo.
01 · Loop
An autonomous loop
Scheduler · watchlist · auto-trader maps ratings to paper orders · allow-list filtering · per-ticker cooldowns · sector caps · cash checks · position manager evaluates open positions every 60s for TP / SL / max-hold. Append-only audit logs.
02 · Brokers
Multi-broker abstraction
Three modes: local Python broker (yfinance fills, JSON-persisted) · Alpaca paper-trading adapter · “shadow” mode running both in parallel with divergence view. Real Alpaca live endpoints are hard-refused at multiple layers.
03 · Dashboard
A local web dashboard
FastAPI backend · React via CDN, no Node toolchain · SVG equity curve · rolling-peak drawdown · win-rate by rating / ticker / model · exit-reason breakdown · LLM cost vs realised P&L joined by run ID. Runs locally; nothing sent to a cloud service.
04 · Codex
Codex OAuth
Runs the engine on a ChatGPT Pro subscription via the Codex backend. LLM cost floor effectively zero if you already have ChatGPT Pro. Token stored encrypted locally. Falls back to the regular OpenAI API if you’d rather pay per token.
05 · Alerts
Multi-channel alerts
Slack · Discord · SMTP email · configurable filter on rating events and order fills · append-only history kept locally. Webhook URLs masked in API responses so a screenshot can’t accidentally leak credentials.
06 · MCP
MCP plug-ins
Registry for adding Anthropic Model Context Protocol servers (Kensho · Aiera · FactSet · Morningstar · LSEG) as analyst tools. Plug-ins advertise category (fundamentals · news · market data · social) · probe endpoint tests credentials.
Honest-by-design touches: every generated report prepends “Research, not advice” and appends a footer with version, commit, provider, models used, run ID, and cost. Closed trades carry the same metadata. 520+ passing unit tests across engine, services, and HTTP routes. The intent: when the system loses money, the journal makes it impossible to pretend it didn’t.
FIG. 04 — HONEST PRIORS · BEFORE RUNNING THIS IN ANGER
Three priors stated before the data starts arriving
The bias of the project: when the data says no, the dashboard says no, the article says no
1
It might fail too. LLMs are not oracles, and a sophisticated framework around language-model outputs does not change the underlying error rate of the model. Sample is still everything. The framework’s outputs are subject to the same statistical noise as any prediction system over small samples.
Highest likelihood
2
If it appears to work, the most likely explanation is variance. The same trap that caught the first article’s candidate edge applies here. A high win rate over fifty trades means much less than it looks. Without out-of-sample confirmation, a flattering early sample tells you almost nothing about whether the system has real edge.
Second-most likely
3
If it appears to work for the right reasons — empirical win rate matches stated confidence, and alpha-versus-benchmark persists across non-overlapping samples — that would be a meaningful research finding. Whether that happens, I don’t know. The point of putting it in the open is that the data will say.
Genuinely open
This is explicitly not a launch announcement for a product anyone should connect a real brokerage account to. The Alpaca live endpoints are hard-refused at multiple layers in the code, and the design choice is deliberate. The right next step is data, not deployment. The bias of the whole project is straightforward: when the data says no, the dashboard says no, the article says no, and no one tries to retroactively rescue the thesis. That’s the contribution.
FIG. 05 — WEEK THREE · WHAT THE METHODOLOGY WILL MEASURE
Four concrete measurements before publishing findings
The hope: write the week-three article from a position of “here’s what the data says”. The fear: another candidate falsified at higher sample. Both outcomes are publishable.
M1 · Sample discipline
Small watchlist for a few weeks before publishing
A handful of tickers across two or three sectors. Long enough to gather sample, narrow enough to keep attention on what’s actually happening per agent. Avoid the noise of a 65-ticker autonomous loop until the smaller version has been read carefully.
M2 · Calibration view
Stated confidence vs. realised win rate
When the system says “75% confident”, do the trades actually win 75% of the time? Same measurement applied to Polybot’s fair-value model. If the model is systematically over-confident, that bias dominates everything downstream.
M3 · Cost accounting
Cost per ticker · per rating · per profitable trade
With Codex OAuth the marginal LLM cost is effectively zero. With the public OpenAI API, each run is hundreds of agent turns. The honest question: does this scale economically if you ever did run it at real cost?
M4 · Non-overlapping windows
Alpha vs benchmark · out-of-sample
Not within-sample alpha — trivially inflatable. Hold out one period entirely, run the system on the next, then check whether the held-out result matches the in-sample stats. If they diverge sharply, the in-sample was curve-fit.
Open under Apache-2.0 with upstream cited from every relevant surface. Not open: the operator’s running results, the specific watchlist, the per-agent prompt customisations, the alert channels, the trade journals — kept local for the same reason Polybot’s per-experiment data is kept local. Publishing exact configurations encourages people to copy them with real money, which is the opposite of what an honest research project should do. Summary findings will be published. Recipes will not.
The bet is on a different mechanism, not a different parameter setting. The point is not to find a money-printing AI. The point is to put honest measurements of these systems into the public record — so the next person looking at the space starts a step further along than the last.
Thorsten Meyer AI · Introducing Forezai · TradingAgents · § 03

Potential of Multi-LLM Committees in Trading

This development matters because it explores a novel approach to AI-driven trading decision-making, moving beyond single-model predictions to collaborative reasoning among specialized LLMs. If successful, it could inform future research on AI team decision processes, with implications for automated trading and financial analysis. While the system currently operates in a simulated environment, advancements here could eventually influence real trading strategies, provided robustness and reliability are demonstrated.

Paper Trading Boss Battles — Beginner Mode: Stock Market for Beginners: A Gamified Guide to Paper Trading and Investing for Teens

Paper Trading Boss Battles — Beginner Mode: Stock Market for Beginners: A Gamified Guide to Paper Trading and Investing for Teens

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Advances in AI and Algorithmic Trading Frameworks

The use of AI in trading has historically focused on rule-based algorithms and machine learning models trained on historical data. Recent efforts, including TauricResearch’s TradingAgents, have experimented with multi-agent systems that articulate reasoning explicitly. Forezai’s fork builds on this foundation, integrating operational features to facilitate research and testing of AI team decision-making in market simulations. Prior research has shown parametric strategies often fail in live conditions, prompting interest in more complex, reasoning-based AI approaches.

“TradingAgents aims to test whether a committee of specialized LLMs can produce decision quality comparable to, or better than, random chance in simulated trading environments.”

— Thorsten Meyer, Forezai developer

AI-Assisted Trading for Retail Traders: A Disciplined Framework for Probability, Risk Control, and Structured Decision-Making

AI-Assisted Trading for Retail Traders: A Disciplined Framework for Probability, Risk Control, and Structured Decision-Making

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Uncertainties About AI Decision Effectiveness

It remains unclear whether the multi-LLM committee can consistently outperform random decision-making in real or simulated trading environments. The system is designed for research rather than live trading, and its effectiveness in generating profitable or even stable decisions has yet to be demonstrated at scale. Additionally, questions about the system’s robustness, sensitivity to data inputs, and potential for overfitting are still open.

Advanced Stock Market Analysis Dashboard: Design and Implementation of a Real-Time Financial Analytics Platform

Advanced Stock Market Analysis Dashboard: Design and Implementation of a Real-Time Financial Analytics Platform

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps in Testing and Evaluation

Forezai plans to run extended simulations of TradingAgents, collecting performance data over various market conditions. Further development will include refining the decision-making algorithms, enhancing the dashboard for detailed analysis, and exploring integration with real trading platforms under strict safeguards. Results from these tests will determine whether this approach warrants further research or potential deployment in live settings.

Algorithmic Trading with Python: Build, Backtest, and Automate Strategies with Code, Data, and Real-World Market Tools

Algorithmic Trading with Python: Build, Backtest, and Automate Strategies with Code, Data, and Real-World Market Tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Can TradingAgents make profitable trades?

Currently, TradingAgents operates in a simulated, research environment. Its ability to generate profitable trades has not been established and is part of ongoing testing.

Does this system predict market movements?

No, the system does not predict market movements directly. Instead, it employs a committee of LLMs to evaluate data and generate trading decisions based on structured reasoning.

Is this system ready for live trading?

No, Forezai emphasizes that the system is for research and simulation only. It includes safeguards to prevent accidental live trading with real money.

What advantages does a multi-LLM committee offer?

The approach aims to leverage diverse reasoning biases and explicit argumentation to produce more robust and transparent decision-making compared to single-model predictions.

Will this approach replace traditional trading algorithms?

It is too early to say; current efforts are experimental. The goal is to understand whether AI team reasoning can improve decision quality, which may inform future algorithm design.

Source: ThorstenMeyerAI.com

You May Also Like

Open Source Boom: Why Tech Companies Embrace Open Code

Harnessing open source code drives innovation and collaboration for tech companies, but the full impact reveals surprising advantages worth exploring.

Climate Tech 2025: Innovations Helping Fight Climate Change

Just as climate innovations accelerate, discover how these breakthroughs could transform our fight against climate change.

Rebrandable client delivery dashboard for AI agencies

A boutique AI agency plans to test a white-label client delivery dashboard, aiming to improve client trust and streamline project updates.

The Defender’s Counter-Cascade.

On May 11, 2026, Google disclosed the first confirmed real-world use of an AI-built zero-day exploit, highlighting deployment gaps in AI-driven security.