canli-mcp / The desk, built into canli-mcp

The chance a company lands above its own guidance

A Canli model: the chance a company's next reported EPS or revenue lands above, within or below its own guided range, from its record against its guidance blended with the market's base rate. Validated walk-forward on S&P 500 guidance (2023 to 2026): it beat the base rate out of sample for EPS and revenue, with the intervals stated. Not a forecast of the price, and not advice.

The tool is beat_probability 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

As the server declares them
ArgumentRequiredType
companyyesstring
measurenoeps|revenue
filed_beforenodate

How a model reaches it

Ask in plain words. The model finds beat_probability 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": "beat_probability",
  "arguments": {
    "company": "<string>"
  }
}

Words that find it

Besides its name, title and description, find_tool matches a question to beat_probability on these words: beat, guidance, probability, earnings, surprise, model, forecast, above, range, sandbag, conservative, management, credibility.

More

Nothing here is investment advice.