classify_textLABSCO SUMMARY
Classification that runs locally and returns confidence, not a guess.
Static embeddings mean the work happens on your machine with nothing sent out, and the categories are yours to define — so the labels match your taxonomy rather than a fixed list someone else chose.
What it is
A text classifier built on Model2Vec static embeddings: classify one text or many, and manage the category set the classifier works against.
What you get
- A single text classified, with confidence scores attached
- Many texts classified in one call
- Ten categories out of the box — technology, business, health, sports, entertainment, politics, science, education, travel and food
- Your own categories added, one at a time or in a batch, each with a description
- The current category list read back, and categories removed when they stop being useful
- Category list and model information also available as resources
Requirements
Python 3.10 or newer and the uv package manager. Nothing to compile: you clone the repository, and uv resolves the dependencies declared inline in the script when it first runs. It speaks stdio locally, and HTTP/SSE or Streamable HTTP when you want it remote.
Setup effort
Clone and run — clone the repository, then add the configuration for your client; uv runs the script from your clone, with nothing to build
USE CASES
01
Routing incoming messages to the right queue
Classify each one against your own categories and act on the confidence score rather than the label alone.
02
Tagging a backlog you never got around to
Send the whole set through in one call instead of one at a time.
03
Teaching it your own taxonomy
Add your categories with descriptions, in bulk if you already have the list written down.
04
Keeping the category set honest
Read back what is currently defined and remove the ones that never get used.
6 tools
Listed in project docs. Your client may show a different list.
batch_classifyadd_custom_categorybatch_add_custom_categorieslist_categoriesremove_categoriesBEFORE YOU INSTALL
What to know first
Tools you get
6 tools named in the project's documentation.
Requires
- Python 3.10+
- uv on PATH
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
Clone the repository, then add the configuration for your client; uv runs the script from your clone, with nothing to build.
Server setup
TERMINAL
git clone https://github.com/baobab-tech/mcp-text-classifier.gitMCP Clients
Claude Desktop — stdio
CLAUDE DESKTOP — STDIO · JSON
{
"mcpServers": {
"text-classifier": {
"command": "uv",
"args": [
"run",
"<YOUR_PATH_TO_MCP_TEXT_CLASSIFIER>/text_classifier_server.py"
]
}
}
}<YOUR_PATH_TO_MCP_TEXT_CLASSIFIER> is the full path of your clone, so the path ends in mcp-text-classifier/text_classifier_server.py. uv fetches the script's dependencies the first time it runs.
Claude Code
CLAUDE CODE · TERMINAL
claude mcp add text-classifier -- uv run <YOUR_PATH_TO_MCP_TEXT_CLASSIFIER>/text_classifier_server.py<YOUR_PATH_TO_MCP_TEXT_CLASSIFIER> is the full path of your clone, so the path ends in mcp-text-classifier/text_classifier_server.py. uv fetches the script's dependencies the first time it runs.
