canli-mcp / Quant finance

Parameter sweep with deflated Sharpe

Run one strategy recipe over a parameter grid (up to 400 variants), rank the variants by Sharpe, deflate the best one for the number tried and give the probability of backtest overfitting (CSCV), so tuning cannot pass off luck as skill.

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

Arguments

As the server declares them
ArgumentRequiredType
recipeyesma_crossover|time_series_momentum|breakout|zscore_reversion|...
argumentsyesobject
gridyesobject

How a model reaches it

Ask in plain words. The model finds strategy_sweep 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": "strategy_sweep",
  "arguments": {
    "recipe": "<ma_crossover|time_series_momentum|breakout|zscore_reversion|...>",
    "arguments": "<object>",
    "grid": "<object>"
  }
}

Words that find it

Besides its name, title and description, find_tool matches a question to strategy_sweep on these words: parameter, sweep, grid, search, optimization, overfitting, deflated, sharpe, multiple, testing, backtest, robustness, pbo, cscv, probability, of.

More

Nothing here is investment advice.