canli-mcp / Markets

Filing tone (local FinBERT)

The tone of a filing or any text, sentence by sentence, from FinBERT run on this machine: an 8-K's press release by default, or a 10-K section. Needs the optional local models (canli-mcp models install). Tone, not truth, and no evidence it predicts a price; the model is licensed for non-commercial use.

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

Arguments

As the server declares them
ArgumentRequiredType
textnostring
companynostring
accessionnostring
formnostring
documentnostring
sectionnostring
max_sentencesnointeger

How a model reaches it

Ask in plain words. The model finds filing_sentiment 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": "filing_sentiment",
  "arguments": {}
}

Words that find it

Besides its name, title and description, find_tool matches a question to filing_sentiment on these words: finbert, sentiment, tone, deep, learning, transformer, local, model, 8-k, press, release, positive, negative, nlp, machine.

More

Nothing here is investment advice.