Introducing Forezai · TradingAgents — a committee of LLMs decides paper-trades

📊 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 launched TradingAgents, a framework where a committee of large language models (LLMs) collaboratively generate paper-trading decisions. This development aims to explore AI’s potential in financial decision-making beyond traditional parametric strategies, emphasizing structured reasoning and multi-agent debate.

Forezai has launched a new version of the TradingAgents framework, enabling a committee of large language models (LLMs) to autonomously generate paper-trading decisions through structured debate and reasoning. This development marks a significant step in AI research aimed at understanding whether multi-agent LLM systems can outperform random decision-making in simulated trading environments.

The new project, called Forezai · TradingAgents, is a fork of an existing open-source framework that orchestrates multiple specialized LLMs to analyze market data and argue their respective positions. Unlike simple prediction models, this system involves thirteen distinct agent roles, including analysts focused on market structure, news, fundamentals, and social sentiment, as well as debate and risk assessment modules. The framework culminates in a final trading decision, expressed as a five-tier rating with price targets and time horizons.

Forezai’s implementation adds operational features such as an autonomous daily scheduler, paper trading via multiple broker interfaces (including Alpaca), position management, and a web dashboard for monitoring. Crucially, it runs entirely locally, with no risk of real-money trading unless explicitly overridden by the operator. The system is designed solely for research and testing, not live trading.

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 Impact on AI-Driven Market Research

This development is significant because it explores whether structured, multi-agent reasoning among LLMs can produce decision-making comparable or superior to random or rule-based strategies in simulated trading environments. If successful, it could influence future AI applications in financial analysis, emphasizing explicit reasoning and debate rather than prediction alone. However, it remains a research tool, and its effectiveness in real markets is unproven.

Amazon

paper trading platform for stocks

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on AI and Trading Strategy Testing

Previous research by Thorsten Meyer and the TauricResearch team involved testing multi-strategy paper-trading bots like Polybot against prediction markets, revealing that many parametric strategies fail to survive real-world data, often appearing profitable in backtests but collapsing in live conditions. This highlighted the limitations of explicit rule-based approaches and raised questions about whether less rule-bound, reasoning-based AI systems could perform better.

The TradingAgents framework was developed to test this hypothesis by having multiple specialized LLMs argue their analyses, providing a more transparent and reasoned decision process. Forezai’s fork enhances this by adding operational features, making it suitable for systematic research rather than just theoretical exploration.

“The core idea is to see if a committee of LLMs, each with different biases and roles, can produce decisions that are at least no worse than random, given the same data a human would see.”

— Thorsten Meyer

Amazon

AI trading decision software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Uncertainty About Real-World Effectiveness

It is not yet clear whether the reasoning and debate among LLMs will translate into meaningful trading advantages in live markets. The system currently operates only in simulated, paper-trading environments, and its performance in real trading conditions remains untested. Additionally, the impact of model biases, data quality, and system robustness are still unknown.

Amazon

multi-agent LLM trading system

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps for Research and Development

Future work will involve extensive backtesting and live simulation to evaluate the decision-making quality of the TradingAgents system. Researchers plan to analyze the correlation between the system’s recommendations and actual market movements, as well as refine agent roles and reasoning strategies. Further, efforts will focus on integrating more sophisticated risk management and expanding the framework’s operational capabilities for broader testing.

Amazon

automated stock analysis dashboard

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Can Forezai TradingAgents be used for real trading?

No, the current system is designed solely for research and paper trading. It does not operate with real money unless deliberately configured and overridden by the user, which is not recommended without thorough testing.

How does the multi-agent system improve decision-making?

The system forces multiple specialized LLMs to analyze data and argue their positions, promoting explicit reasoning and debate. This contrasts with single prediction models and aims to reduce biases and improve robustness.

What are the main limitations of this approach?

The primary limitations include the lack of proven effectiveness in live markets, potential biases within LLMs, and the challenge of translating simulated decision-making into real-world gains. The system is still experimental and primarily for research purposes.

Will this technology replace human traders?

Currently, it is a research tool aimed at understanding AI decision-making processes. There is no indication that systems like Forezai TradingAgents will replace human traders but may complement analysis and decision support in the future.

Source: ThorstenMeyerAI.com

This content is for general information only and is not financial, tax or legal advice. Consult a qualified professional for decisions about your money.
You May Also Like

ShinyHunters · The New APT Model.

ShinyHunters has evolved into a scalable, AI-enabled extortion collective operating as a distributed brand with a new operational model, surpassing traditional APTs.

Kyber (YC W23) Is Hiring a Founding Marketer

Kyber, a YC-backed enterprise AI platform, is seeking a Founding Marketer to build its content and community efforts, signaling a new phase of growth.

Iconic CA Burger Joint To Close 10 Restaurants And Sell Dozens More, Blaming $20 Minimum Wage

A major California franchise operator is planning to close 10 restaurants and sell dozens more, citing financial losses linked to the $20 minimum wage for fast-food workers.

White-collar professional services. The Tier 1 displacement.

Major firms in legal, banking, consulting, and accounting are reducing graduate intake and testing AI tools, signaling structural change in professional services.