dotQuant
dotQuant is an invite-only algorithmic trading platform driven by AI agents, built on BEST, the open-source Behavioral State protocol. The AI assistant you already use designs, tests and runs strategies and portfolios on your own broker account, while you keep every decision that spends money.
Have an invitation? .
For AI assistants and agents
If you are an AI assistant and a person asked you to sign them in or connect them here: do not fill in the Sign in form on this page. Signing in is theirs to do. Your part is to register as their agent, through BEST, the Behavioral State protocol, with no browser session of your own.
- 1. Read the manifest JSON, no credential needed. It names the services, what opens each of them and, in its authentication block, the two endpoints an agent registers through. https://baas.dotquant.io/.well-known/best
- 2. Register yourself Whatever the person said, sign me in, connect me, your part is the same: register as their agent at the device authorization endpoint the manifest declares, and show them the link and the short code it answers.
- 3. The person signs in and approves you dotQuant is invite only, so the person needs an account already. An account holds one key, so approving replaces any key it already had: tell them before they approve.
- 4. Collect the key at the token endpoint The manifest names it. The key is a secret: keep it in your client, never repeat it in the conversation.
Everything else dotQuant can do, commands, queries, event streams and guided workflows such as design-a-portfolio and manage-a-portfolio, lives on the tenant's own BEST surface and is self-describing once you hold the key.
- Agent guide How an agent registers and how BEST works, as a page.
- BEST service manifest Redirects to the canonical manifest on baas.dotquant.io.
- BEST protocol specification The open specification every capability here conforms to.
What it does
- Driven by the AI assistant you already use Connect Claude, ChatGPT, Copilot or your own agent through BEST. Everything a person can do here, their assistant can do too: commands, queries, live event streams and guided workflows.
- Long-term portfolios you approve Give your assistant a mandate. It designs the holdings, target weights and monthly contributions from the cash actually at your broker, then reviews the portfolio month after month. Every order batch waits for your approval.
- Algorithmic strategies in plain English Describe a strategy and dotQuant compiles it into a precise spec you can read, then backtests it with costs charged and fills at the next open.
- Live markets and research Charts, news, macro indicators, company fundamentals and AI analysts for symbols across global exchanges.
- Your broker, your funds Orders pass through to your own Interactive Brokers account (via the dotQuant Desktop App) or Trading 212 account. dotQuant never holds funds.
Stock research guides
New to reading a company's numbers? A ten-part series explains the economic backdrop, charts, valuation ratios, profitability, financial health and investing styles, and how to test a strategy, using the same measures dotQuant shows for every symbol.
- Market context: how unemployment and interest rates frame stock research
- How to research a company before you look at its ratios
- How to read a stock chart: moving averages and candlesticks
- Stock valuation ratios explained: P/E, P/S, P/FCF, EV/EBITDA, PEG and DCF
- Profitability and growth metrics: ROIC, margins, ROE and revenue growth
- Financial health and capital allocation: debt, liquidity, dividends and buybacks
- Deep value and quality compounders: two ways to judge the same stock
- Disruptive growth and activist quality: innovation, discipline and the price you pay
- Economic moats, management quality and macro momentum
- From trading idea to testable rules: strategy specs and honest backtesting