MALI is an AI stock research and market-intelligence terminal for analysts, portfolio managers and investment teams. Firms can take the Institutional plan (custom AI agents, white-label and SSO built to order), and a public REST API (OpenAPI 3.1) and an MCP server let teams bring MALI's research into their own tools.
It brings prices, SEC filings, economic data and news into one place and puts AI agents to work on them, and it is built to show its work: where each figure came from, what an agent argued and why, and how its calls have scored.
What sets it apart
• Multi-agent debate: six agents (Bull, Bear, Macro, Quant, Sentiment and Risk) read the same evidence on a stock, cross-examine each other's weakest claims, and a judge works out where the weight of the argument lands.
• Forecasts as ranges: the forecast engine gives probability ranges over a horizon, not a single price target.
• A public track record: every call is hash-chained when it is made and scored against SPY when it comes due, with Brier scores, calibration and 95% intervals. Below 30 scored calls it draws no conclusion. Anyone can download every call as CSV.
What else it does
• SEC EDGAR financial statements and 13F holdings, read straight from the filings
• Market regimes, a cross-market contagion map and FinBERT news sentiment
• A live market scanner and a plain-English stock screener
• Options strategy generator, backtesting and a scenario lab
• Portfolio tracking with transaction import from Schwab, Fidelity, Vanguard, Robinhood and Interactive Brokers
• A daily AI brief, event alerts and a voice assistant
How it's built
When a source can't be read, MALI says so instead of filling the gap with an estimate. It doesn't connect to brokerage accounts or place orders, and nothing on it is investment advice.
Plans: Starter is free (10 AI queries a day). Pro is $79 a month and Edge $95 a month, each with a 14-day trial on a first subscription; annual billing is $790 and $969 a year. Institutional is custom.