Multi-agent AI research for any ticker

Analysts, bull/bear researchers, a trader, and a risk team debate every call — then deliver one clear buy/sell/hold recommendation, not just a pile of data. For stocks, ASX/LSE/HKEX and 12 other exchanges, and crypto.
See pricing
Research opinion, not financial advice — see disclaimer below.
Sample output
NVDA 2026-09-25
Bull agent Data-center demand keeps outrunning supply — margins hold even as capex scales.
Bear agent Valuation already prices in three years of flawless execution. Any air pocket gets punished hard.
Recommendation HOLD
Illustrative example — every real run shows the full reasoning trail.

How the analysis is made

Five stages, each a separate LLM agent, each free to disagree with the last.

1

Data fetch

Fundamentals, Sentiment, News, and Technical agents each call live data tools for your ticker and date.

2

Bull vs. bear

Two researcher agents argue opposing cases. One-sided or thin-sample crowd sentiment is flagged as a contrarian caution, not confirmation.

BullMargins hold up.
BearValuation is stretched.
3

Proposal

A trader agent turns the debate into a proposed action — sizing, entry, and stop.

4

Risk stress-test

Aggressive, conservative, and neutral risk agents pressure-test the proposal.

5

Executive verdict

A portfolio manager agent issues the final recommendation — sometimes overriding an earlier stage.

How current is the data, and how much should you trust it?
  • Data is pulled fresh each run for the date you specify — as current as the underlying providers (mainly Yahoo Finance) allow, not real-time.
  • It's not the LLM's own training knowledge — every run calls live tools.
  • Coverage can be thinner for non-US tickers.
  • Crowd sentiment (Reddit, StockTwits) gets a built-in contrarian check — markets can be irrational, so a thin sample or a near-unanimous bull/bear split is flagged as a caution sign, not confirmation, before it reaches the bull/bear debate.
  • LLM output is stochastic and this hasn't been backtested for accuracy — treat every recommendation as one research opinion to sanity-check, not a validated signal.

Pricing

Simple monthly plans, capped so cost stays predictable.

Free trial
$0
  • 3 analysis runs, no card required
  • Stocks, 15 global exchanges, and crypto
  • One trial per email address
Starter
$19/month
  • 15 analysis runs per month
  • Stocks, 15 global exchanges, and crypto
  • Full stage-by-stage reasoning breakdown
Pay per ticker
$2.99 / ticker
  • No subscription — pay once, use anytime
  • Choose 1 run or a 12-run pack below
Already have an access token? Sign in.

Frequently asked questions

What is TradeRight?

TradeRight is a multi-agent AI trading research tool. It runs a structured debate — fundamentals, sentiment, news, and technical analyst agents, then bull vs. bear researcher agents, a trader agent, and a risk team — and delivers one clear buy/sell/hold recommendation for a stock or crypto ticker, showing the reasoning from every stage.

Is TradeRight financial advice?

TradeRight always gives you a clear call — one buy/sell/hold recommendation from the portfolio-manager agent, not just a pile of data. That said, it's not financial advice: it's a simulated opinion from an LLM multi-agent debate over public market data, TradeRight does not place trades, its accuracy has not been independently validated, and outputs can differ between runs on the same ticker and date.

How does the multi-agent debate work?

Five stages, each a separate LLM agent free to disagree with the last: an analyst team (fundamentals, sentiment, news, technical) each calls live data tools for the ticker and date; two researcher agents argue the bull and bear case; a trader agent turns the debate into a proposed action with sizing, entry, and stop; an aggressive/conservative/neutral risk team stress-tests that proposal; and a portfolio manager agent issues the final recommendation, sometimes overriding an earlier stage.

What markets and tickers does TradeRight cover?

Stocks across 15 global exchanges — including NYSE/NASDAQ, ASX, LSE, HKEX, TSX, Euronext Paris, and XETRA — as well as crypto tickers such as BTC-USD and ETH-USD.

How current is the data TradeRight uses?

Data is pulled fresh each run for the date you specify, mainly from Yahoo Finance and news/sentiment sources — it is not real-time and not the underlying LLM's own training knowledge. Coverage can be thinner for non-US tickers.

How much does TradeRight cost?

Starter is $19/month for 15 analysis runs. Pro is $49/month for 50 analysis runs. Both cover stocks across 15 global exchanges and crypto, with a full stage-by-stage reasoning breakdown.

Who makes TradeRight?

TradeRight is built by Australscience, a Sydney-based deep-tech AI research company, on top of the open-source TradingAgents multi-agent framework.

Disclaimer & methodology notes

Disclaimer: Not Financial Advice

The information, tools, data, and content provided through TradeRight are for general informational and educational purposes only. They do not constitute financial, investment, trading, tax, or legal advice, and should not be relied upon as such.

TradeRight does not consider your personal objectives, financial situation, or needs. Before acting on any information, you should consider whether it is appropriate for your circumstances and, where necessary, seek advice from a licensed financial adviser.

Trading and investing involve significant risk. The value of investments can go down as well as up, and you may lose some or all of the money you invest. Past performance is not a reliable indicator of future results.

Any market data, analysis, signals, or opinions provided through TradeRight may be incomplete, delayed, or inaccurate, and are provided without warranty of any kind. TradeRight, its owners, employees, and affiliates accept no liability for any loss or damage arising from the use of, or reliance on, this product or its content.

All decisions to buy, sell, or hold any financial product are made solely at your own discretion and risk.

Methodology notes