# angrysky56/AsterMind-ELM MCP Server

Source: https://glimind.com/server/glama%2Fangrysky56%2Fastermind-elm-mcp-server
Registry id: `glama/angrysky56/astermind-elm-mcp-server`
Status: unknown

## Tools (16)

- `delete_model`: Delete a model from memory.
- `detect_drift`: Detect distribution drift between prediction windows using Jensen-Shannon divergence.
- `generate_embedding`: Generate model-specific hidden-feature vectors for input text.
- `get_confusion_matrix`: Get the confusion matrix for a model based on logged predictions.
- `get_model_metrics`: Get accuracy, latency, label counts, and distribution drift metrics for a model.
- `list_model_versions`: List all versions of a model in persistent storage.
- `list_models`: List all models currently in memory with metadata.
- `load_model_persistent`: Load a model from persistent storage.
- `load_training_dataset`: Load a training dataset from persistent storage.
- `predict`: Predict labels for input text and return ranked confidence scores.
- `save_model`: Save an in-memory model summary.
- `search_similar`: Search stored vectors by cosine similarity.
- `store_embeddings`: Store arbitrary equal-dimension vectors for similarity search.
- `store_model_persistent`: Store a model persistently in SurrealDB.
- `store_training_dataset`: Store a training dataset persistently.
- `train_classifier`: Train a multi-class text classifier from labeled examples.

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.