Momentum stays intact if guidance confirms durable AI spend.
Explainable multi-agent market scenario debate
Debate a trade setup before it becomes a journal entry.
Enter a ticker or market theme and preview how a bull agent, bear agent, and risk officer challenge the same scenario. Convert the debate into probability, invalidation points, and a paper-trade note.
Example: Brazilian bank earnings, India IT services rerating, Gulf energy infrastructure, or a US mega-cap catalyst.
watchlist > debate > probability > journal
TradingAgent Sim
US equities
research
Multiple expansion is fragile if revenue quality looks concentrated.
Size down until event drift and liquidity gaps are reviewed.
58%
base scenario probability
5
watchlist slots on Pro
80
briefs per month
Invalidation points
- Guidance fails to confirm the next growth leg.
- Volume fades below the prior breakout range.
- Macro rates shock lifts discount-rate pressure.
Research workflow
Three agents, one inspectable scenario.
The product starts as a research-only workflow: debate the setup, write the invalidation checklist, and export the journal before any broker-neutral paper tracking is added.
Bull agent
Builds the strongest upside case, names catalysts, and identifies evidence that would improve the probability.
Bear agent
Challenges valuation, crowding, liquidity, data quality, and story risk so the setup does not become a one-sided note.
Risk officer
Forces invalidation points, sizing discipline, review timing, and paper-trade journal fields into the workflow.
Signal debate transcript
Readable enough for a research desk, structured enough for automation.
Each debate separates claims, objections, missing evidence, probability drivers, and invalidation points. The transcript can be turned into a watchlist research report or a paper-trade journal entry.
1. Scenario brief
2. Agent debate
3. Probability
4. Invalidation
5. Journal export
Paper-trade journal
Research stays useful when the scenario can be revisited.
Export the watchlist, transcript, probability, thesis, invalidation points, and review date into a paper-trade journal. The workflow is broker-neutral and does not route orders.
ProbabilityRecord the base case and the conditions that would change it.
InvalidationName the exact facts, levels, or events that break the thesis.
ReviewSchedule a follow-up after catalysts, earnings, macro data, or risk events.
Pricing
Pro annual is selected by default.
Annual billing is 50% less than paying month to month. Pricing is based on watchlists and research reports.
Analyst
Single researcher
$24.50/ mo
Billed annually as $294. 50% annual savings.
1 watchlist and 15 research briefs
- Bull, bear, and risk-officer debate preview
- Scenario probability notes
- Invalidation checklist
- Paper-trade journal export
Pro
Active research workflow
$99.50/ mo
Billed annually as $1194. 50% annual savings.
5 watchlists and 80 research briefs
- Everything in Analyst
- Signal debate transcript
- Queue jobs for recurring scenarios
- Broker-neutral paper tracking
- LLM-ready research report export
Desk
Research desk
$299.50/ mo
Billed annually as $3594. 50% annual savings.
25 watchlists and shared reports
- Everything in Pro
- Desk-level scenario library
- API access and custom queue schedules
- Review roles and report approvals
- Priority support
Research guides
Practical pages for AI trading research workflows.
AI trading agentsUse AI trading agents to debate market scenarios, compare bullish and bearish evidence, define invalidation points, and export paper-trade journals without trade execution.multi-agent trading AIRun multi-agent trading AI debates that separate bullish evidence, bearish objections, and risk controls before a research note is finalized.AI trading simulationUse AI trading simulation to rehearse scenarios, identify invalidation points, and maintain paper-trade journals before any live brokerage workflow is considered.trading agents AITrading agents AI can help analysts document scenario debates, risk controls, invalidation points, and paper-trade journals for research workflows.