canli-mcp / The desk, built into canli-mcp
Stress a portfolio across thousands of synthetic worlds
Run a portfolio through thousands of synthetic worlds: real market history resampled in blocks, with real crises (2008 banks, 2022 chips, 2020 crash, others) dropped in at random times and severities, alone or combined, or your own shocks. Gives each scenario's range of returns, tail loss and drawdown, the holdings that hurt most and how far the answer can be trusted. Scenarios, not forecasts.
The tool is stress_worlds in canli-mcp 0.22.0, free and MIT-licensed, with no account and no API key. It is part of canli-mcp itself, so it is there as soon as canli-mcp is added.
Arguments
| Argument | Required | Type |
|---|---|---|
holdings | yes | {symbol,weight}[] |
scenarios | no | string[] |
custom | no | {name,moves,days}[] |
worlds | no | integer |
horizon_days | no | integer |
seed | no | integer |
How a model reaches it
Ask in plain words. The model finds stress_worlds 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": "stress_worlds",
"arguments": {
"holdings": "<{symbol,weight}[]>"
}
}Words that find it
Besides its name, title and description, find_tool matches a question to stress_worlds on these words: stress, test, scenario, simulation, synthetic, macro, monte, carlo, crisis, what, if, tail, risk, drawdown, antifragile, banking, compute, shock, portfolio, resilience, worlds.
More
Nothing here is investment advice.