📊 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 fork of a multi-LLM research framework that automates paper trading using a committee of specialized AI agents. The system aims to evaluate AI decision-making in simulated markets, with operational features for research and testing.
Forezai has released a new version of its TradingAgents framework, adding operational automation to a multi-LLM research system designed for simulated trading. This development enables researchers and developers to run autonomous paper trades based on AI-generated decisions, with detailed logging and multi-broker support, marking a significant step toward testing AI-driven trading strategies in a controlled environment.
The TradingAgents framework, originally developed by TauricResearch, employs a committee of thirteen specialized large language models (LLMs) that analyze market data, debate, and synthesize trading decisions. Previously, it was a research tool that generated structured reports and recommendations without operational automation. The new Forezai fork introduces an operational layer, including an automated scheduler, paper trading interface, position management, and multi-broker support, allowing continuous, autonomous trading simulation. It also features a web dashboard for monitoring performance metrics, risk, and decision rationale, all running locally without cloud data exchange. The system explicitly does not trade real money unless operators override safety measures, emphasizing its role as a research instrument rather than a live trading platform.AGENTS
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
Analysts · Debate · Risk · Decision
78% / -33%
Polybot prior: fleet win rate
combined with -33% bankroll
combined with -33% bankroll
520+
Passing unit tests across engine,
services, HTTP routes (starting baseline)
services, HTTP routes (starting baseline)
€0 floor
LLM cost on Codex OAuth
(falls back to public API per token)
(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 AI-Driven Autonomous Paper Trading
The introduction of Forezai’s TradingAgents system represents a notable advance in AI research for trading. By automating the decision-making and operational processes, it allows for rigorous testing of multi-agent AI strategies in a simulated environment, reducing human bias and enabling large-scale experimentation. This development could influence future AI trading research, providing insights into how AI committees reason, debate, and make decisions without relying on single-model predictions. It also highlights ongoing efforts to assess AI capabilities in complex, real-world-like tasks, with potential implications for both research and the future of automated trading technology. However, it remains a research tool; its effectiveness in live markets is unproven and subject to further validation.
Chemical Process Simulation and the Aspen HYSYS Software
- Condition: Used Book in Good Condition
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background of AI Trading Research and Framework Development
The concept of AI-driven trading has long been explored, with early efforts focusing on parametric strategies that often fail to survive real-market conditions. TauricResearch’s TradingAgents project, published earlier this year, introduced a multi-agent system where specialized LLMs analyze market data, debate, and synthesize trading recommendations. This approach emphasizes explicit reasoning and structured debate among AI agents, contrasting with traditional single-model predictions. Prior to the Forezai fork, the framework was limited to research and testing without operational automation, focusing on understanding AI reasoning and decision-making in simulated environments. The recent release expands its capabilities, enabling continuous, autonomous paper trading and detailed performance tracking, thus bridging the gap between research and operational testing.“The Forezai fork transforms our multi-agent framework into a practical research instrument, allowing continuous simulation and detailed analysis of AI decision-making in trading.”
— Thorsten Meyer, lead developer at TauricResearch
Uncertainties About Real-World Applicability
It is not yet clear how well the AI committee’s decisions in the simulated environment will translate to real markets. The system is designed for research and testing, not live trading, and its effectiveness in actual trading conditions remains unproven. Additionally, the impact of potential biases, decision transparency, and safety measures in live settings requires further evaluation.
Next Steps in Testing and Validation
Researchers and developers will likely conduct extended simulations to evaluate the AI committee’s decision consistency, robustness, and risk management. Future work may include integrating live market data, refining agent roles, and exploring the system’s performance across different asset classes. Validation against historical data and controlled live experiments are expected to follow to assess real-market viability.
Key Questions
Can the Forezai TradingAgents system trade with real money?
No, the current system is designed for paper trading and research purposes only. It explicitly refuses to trade real money unless operators override safety measures, which is not recommended for general use.
How does the multi-agent system make trading decisions?
The system employs specialized LLMs that analyze market data, debate, and synthesize their findings into a final recommendation. It emphasizes explicit reasoning and multi-voice debate rather than single-model predictions.
What are the main operational features added in the Forezai fork?
The fork includes an autonomous scheduler, paper trading interface, position management, multi-broker support, detailed logging, and a web dashboard for monitoring performance and decision rationale.
What are the limitations of this AI trading research framework?
Its primary limitation is that it is a simulated environment, and its decisions have not been proven effective in live markets. Additionally, safety, bias, and decision transparency are ongoing concerns that require further study.
Source: ThorstenMeyerAI.com
