Workspace

Analyze stocks, optimize portfolios, or backtest strategies

Mock Mode

Market research with quant analysis and explainable recommendations.

Stock Analysis

Portfolio Optimization

Backtest Strategy test

QuantBot is built for stock research, portfolio analysis, and strategy validation.

It can fetch market context from internet and API sources, run quantitative analysis, compare multiple holdings, and turn the evidence into a transparent buy / hold / sell style recommendation.

Try prompts like "Analyze NVDA", "Compare CBA and WBC", "Optimize a portfolio with AAPL, MSFT, NVDA", or "Backtest AAPL from 2025-01-01 to 2026-03-18".

QuantBot

Start an analysis

Use plain language or tap a starter below.

Professional Research View

Analysis appears here after each request.

Use the chat panel to investigate a share, test a trading idea, or ask for a portfolio recommendation. QuantBot will surface price structure, quant signals, risk framing, and a plain-English recommendation.

Market context News, sentiment, and macro backdrop

Quant engine Technical, valuation, and confidence scoring

Decision support Explainable recommendation with risks

Single Share

Equity deep dive

Review trend, volatility, sentiment, analyst targets, and signal alignment in one report.

Portfolio

Allocation ideas

Compare multiple names, rank them, and surface diversification-aware positioning.

Backtest

Historical validation

Replay a strategy window to inspect performance, drawdowns, and decision robustness.

Prompt starter

"Analyze ASX:CBA"

Natural-language prompts work well for Australian and US share analysis workflows.

⚠️ Advisory: Multi-Agent Mode (Experimental)

Multi-Agent System Disclaimer

The Multi-Agent Committee execution framework is an experimental feature. Quantitative models and algorithms remain the dominant industry standard for systematic equity research due to critical agent limitations:

1. High Latency Multi-turn debates between bull/bear agents and consensus reviews add significant processing delay.

2. High Token Cost Generating massive qualitative reasoning across 5 distinct LLM steps demands high API usage.

3. Hallucination Risk LLMs can generate ungrounded qualitative justifications or misinterpret precise numeric bounds.

4. Optimism Bias LLMs lean bullish due to sycophancy/data distributions, necessitating rigid quantitative guards.