canli-mcp / Factor backtests

Run a whole paper exchange

Run a whole paper exchange of scripted traders in one call and report it: for comparing set-ups, such as how much leverage turns a shock into a cascade. A simulation, not a forecast.

The tool is exchange_run in canli-mcp 0.22.0, free and MIT-licensed, with no account and no API key. It comes from canli-backtest-mcp, which canli-mcp includes, and it is also its own server.

Arguments

As the server declares them
ArgumentRequiredType
seednointeger
assetsnointeger
personasno{market_maker,value,momentum,noise,deleverager,stop_loss,levered_long}
starting_cashnonumber
maintenance_marginnonumber
leveragenonumber
shocksno{step,kind,asset,percent,steps,quantity}[]
stepsyesinteger

How a model reaches it

Ask in plain words. The model finds exchange_run with canli-mcp's find_tool, reads its arguments with describe_tool if it needs to, and runs it with run_tool. The call has this shape, with your values in place of the angle brackets:

{
  "name": "exchange_run",
  "arguments": {
    "steps": "<integer>"
  }
}

Words that find it

Besides its name, title and description, find_tool matches a question to exchange_run on these words: simulate, ai, agents, trading, against, each, other, simulation, market, cascade, flash, crash, margin, calls, forced, selling, limit, order, book, liquidity, withdrawal, stress.

Recorded questions that use it

Each page shows the call and the full answer it returned, recorded with no API key.

More

Nothing here is investment advice.