canli-mcp / Backtest validation
Check a signal for lookahead
Does a signal look ahead? plan gives cuts; rerun your code on the first cut rows for each, then compare. Rows that changed used later rows; the result names the pattern and horizon. No code is sent. A pass covers only these cuts: the columns are the caller's, and survivorship, vendor revisions or a leak no cut reaches leave no trace.
The tool is check_leakage in canli-mcp 0.22.0, free and MIT-licensed, with no account and no API key. It comes from canli-validation-mcp, which canli-mcp includes, and it is also its own server.
Arguments
| Argument | Required | Type |
|---|---|---|
action | yes | plan|compare |
observations | no | integer |
prefixes | no | integer |
seed | no | integer |
cuts | no | integer[] |
columns | no | object |
columns_file | no | string |
timestamps | no | string[] |
tolerance | no | number |
How a model reaches it
Ask in plain words. The model finds check_leakage 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": "check_leakage",
"arguments": {
"action": "<plan|compare>"
}
}More
Nothing here is investment advice.