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 · TradingAgents has launched a framework where multiple LLMs act as a decision-making committee for simulated trading. This development aims to explore AI’s potential in market decision-making beyond simple prediction, with confirmed operational features and ongoing evaluation.

Forezai · TradingAgents has introduced an operational version of its multi-LLM committee system designed for paper-trading simulations, marking a significant step in AI-driven trading research.

The new system extends the existing TradingAgents framework by adding an autonomous operational layer that manages daily trading cycles, order placement, position evaluation, and logging, all within a secure, simulated environment. It features a multi-broker abstraction supporting local, Alpaca paper, and shadow modes, along with a web dashboard for monitoring performance. The framework does not trade real money unless deliberately overridden, emphasizing research and testing rather than live trading.

This development is built on the premise that a committee of specialized LLMs, each with distinct roles—such as market analysis, debate, risk assessment, and decision synthesis—can produce trading decisions that are at least as effective as random choices, if not better. The system is designed to articulate reasoning explicitly through multiple voices, avoiding reliance on raw data recall, and to generate transparent, structured outputs for each decision cycle.

While the framework is operational, its effectiveness in outperforming simple models or human traders remains under evaluation. The project emphasizes rigorous logging, audit trails, and safety features to prevent unintended real-money trading, aligning with its research focus.

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

Implications of Multi-LLM Trading Committee

This development matters because it explores whether AI, structured as a committee of large language models with specialized roles, can make more nuanced and transparent trading decisions than traditional rule-based or single-model approaches. If successful, it could inform future research into AI-driven decision-making systems in finance, emphasizing explainability and collaborative reasoning.

Although not designed for live trading, the framework’s ability to simulate and analyze complex decision processes offers valuable insights into AI’s potential to understand and navigate market dynamics, potentially influencing future AI applications in finance and beyond.

Chemical Process Simulation and the Aspen HYSYS Software

Chemical Process Simulation and the Aspen HYSYS Software

Used Book in Good Condition

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Evolution of AI in Trading Research

Previous efforts in AI trading focused on parametric strategies, which rely on explicit rules with hand-tuned parameters. These approaches often failed to survive out-of-sample testing, revealing that many perceived edges are mechanical artifacts that vanish under honest evaluation.

The recent shift has been toward more flexible, less rule-bound AI systems, such as multi-agent frameworks utilizing large language models. The TradingAgents project, originally developed by TauricResearch, exemplifies this approach by structuring LLMs into specialized roles that debate, analyze, and synthesize trading signals without promising prediction accuracy. The current Forezai fork builds on this foundation, adding operational features to facilitate research and testing.

“This system provides a structured environment to evaluate whether a committee of LLMs can produce decision-making that is at least as reliable as random choice, with the added benefit of explicit reasoning.”

— Thorsten Meyer, lead developer of Forezai · TradingAgents

The Intelligent AI Investor: A Beginner’s Guide to Using AI Tools for Informed Investment Decisions, Risk Management, and Wealth Building (Trading & Investing Series Book 7)

The Intelligent AI Investor: A Beginner’s Guide to Using AI Tools for Informed Investment Decisions, Risk Management, and Wealth Building (Trading & Investing Series Book 7)

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Unanswered Questions About AI Trading Effectiveness

It is still unclear how well the committee of LLMs will perform in real market conditions over extended periods, or whether their reasoning will translate into consistent, profitable strategies beyond simulated environments. The system is designed for research, and its predictive or trading edge remains to be validated through ongoing testing.

Additionally, the impact of different role configurations, the influence of model biases, and the robustness of the decision process under market stress are still under investigation.

Amazon

multi-LLM trading dashboard

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps in Evaluating AI Committee Trading

The project will continue to run extended simulations, analyze performance metrics, and refine the agent roles and decision protocols. Researchers plan to compare results against baseline models and human strategies, aiming to identify strengths and limitations of the AI committee approach. Further, efforts will focus on enhancing transparency, safety features, and possibly transitioning to live testing with strict safeguards.

Automated Trading with R: Quantitative Research and Platform Development

Automated Trading with R: Quantitative Research and Platform Development

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Can Forezai · TradingAgents trade real money?

No, the current system is designed for paper trading and simulation. Real-money trading requires deliberate overrides and carries inherent risks.

How do the LLMs make trading decisions in this system?

The LLMs act in specialized roles—such as analysts, debaters, risk assessors, and decision synthesizers—and articulate their reasoning explicitly through structured reports and arguments, which are then combined into a final decision.

What advantages does this multi-LLM approach offer over traditional models?

It encourages explicit reasoning, debate among diverse perspectives, and transparency in decision-making, potentially reducing overfitting and mechanical artifacts common in parametric strategies.

Is this system intended for live trading in the future?

Currently, it is a research tool. While future developments may explore live trading, significant safety, validation, and regulatory considerations would need to be addressed first.

What are the main limitations of this approach?

Its effectiveness in real markets is unproven, and the system relies on simulated trading environments. The interpretability and robustness of AI decisions under diverse market conditions remain under study.

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

India: Build the Rails First

Thorsten Meyer AI says India’s welfare strategy rests on Aadhaar, UPI and DBT: thin benefits delivered at huge scale.

The Compounding Error Problem — Why 99.9% Alignment Decays to 60% in 500 Generations

Analysis of how 99.9% alignment accuracy drops significantly over multiple AI generations, raising concerns for recursive self-improvement safety.

The Bubble Question, Disentangled: 1999 vs 2026 Category by Category

A detailed analysis compares the 1999 dotcom bubble with the 2026 AI cycle, highlighting differences in valuation, fundamentals, and risks across categories.

EuroHPC. The compute substrate.

An in-depth analysis of EuroHPC’s compute substrate, its current capabilities, limitations, and implications for Europe’s AI ambitions amid recent developments.