Text Classification (Model2Vec)

A server for text classification using static embeddings from Model2Vec, supporting multiple transports like stdio and HTTP/SSE.

Text Classification MCP Server (Model2Vec)

A powerful Model Context Protocol (MCP) server that provides comprehensive text classification tools using fast static embeddings from Model2Vec (Minish Lab).

This server provides6 essential tools,2 resources, and1 prompt templatefor text classification:

- classify_text- Classify single text with confidence scores
- batch_classify- Classify multiple texts simultaneously

- add_custom_category- Add individual custom categories
- batch_add_custom_categories- Add multiple categories at once
- list_categories- View all available categories
- remove_categories- Remove unwanted categories

- categories://list- Access category list programmatically
- model://info- Get model and system information

- classification_prompt- Ready-to-use classification prompt template

- Zero-install: Justuv run— dependencies are declared inline (PEP 723)
- Multiple Transports: Supports stdio (local), HTTP/SSE, and Streamable HTTP
- Fast Classification: Uses efficient static embeddings from Model2Vec
- 10 Default Categories: Technology, business, health, sports, entertainment, politics, science, education, travel, food
- Custom Categories: Add your own categories with descriptions
- Batch Processing: Classify multiple texts at once
- Resource Endpoints: Access category lists and model information
- Prompt Templates: Built-in prompts for classification tasks

No separate install step needed — dependencies are declared inline in the script (PEP 723) and resolved automatically byuv.

# SSE on default port 8000 uv run text_classifier_server.py --http # SSE on custom port uv run text_classifier_server.py --http 9000

uv run text_classifier_server.py --streamable-http

Add to~/Library/Application Support/Claude/claude_desktop_config.json:

{ "mcpServers": { "text-classifier": { "command": "uv", "args": ["run", "/path/to/text_classifier_server.py"] } } }

Start the server withuv run text_classifier_server.py --http, then add:

{ "mcpServers": { "text-classifier": { "url": "http://localhost:8000/sse" } } }

claude mcp add text-classifier -- uv run /Users/olivier/DEV/mcp-text-classifier/text_classifier_server.py

Classify a single text into predefined categories with confidence scores.

- text(string): The text to classify
- top_k(int, optional): Number of top categories to return (default: 3)

Returns:JSON with predictions, confidence scores, and category descriptions

classify_text("Apple announced new AI features", top_k=3)

Classify multiple texts simultaneously for efficient processing.

- texts(list): List of texts to classify
- top_k(int, optional): Number of top categories per text (default: 1)

Returns:JSON with batch classification results

batch_classify(["Tech news", "Sports update", "Business report"], top_k=2)

Add a new custom category for classification.

- category_name(string): Name of the new category
- description(string): Description to generate the category embedding

add_custom_category("automotive", "Cars, vehicles, transportation, automotive industry")

Add multiple custom categories in a single operation for efficiency.

- categories_data(list): List of dictionaries with 'name' and 'description' keys

Returns:JSON with batch operation results

batch_add_custom_categories([ {"name": "automotive", "description": "Cars, vehicles, transportation"}, {"name": "music", "description": "Music, songs, artists, albums, concerts"} ])

List all available categories and their descriptions.

Returns:JSON with all categories and their descriptions

Remove one or multiple categories from the classification system.

- category_names(list): List of category names to remove

Returns:JSON with removal results for each category

remove_categories(["automotive", "custom_category"])

- categories://list: Get list of available categories with metadata
- model://info: Get information about the loaded Model2Vec model and system status

- classification_prompt: Template for text classification tasks with context and instructions

Returns:Formatted prompt for classification with available categories listed

npx @modelcontextprotocol/inspector uv run text_classifier_server.py

# Manual model download uv run python -c "from model2vec import StaticModel; StaticModel.from_pretrained('minishlab/potion-base-8M')"

- Model:minishlab/potion-base-8Mfrom Model2Vec
- Similarity: Cosine similarity between text and category embeddings
- Performance: ~30MB model, fast inference with static embeddings
- Protocol: MCP specification 2024-11-05
- Transports: stdio, HTTP+SSE, Streamable HTTP
- Fork the repository
- Create a feature branch
- Add tests for new functionality
- Submit a pull request

MIT License - see LICENSE file for details.

- Model2Vecby Minish Lab for fast static embeddings
- Anthropicfor the Model Context Protocol specification
- FastMCPfor the excellent Python MCP framework

A Python-based server that locally indexes codebases using ChromaDB to provide semantic search for tools like Cursor.