llamafarm

Context-aware assistant for building and troubleshooting LlamaFarm projects. Activates for LlamaFarm configuration, llamafarm.yaml, RAG pipelines, API integration, MCP development, debugging, deployment, anomaly detection, text classification, OCR, NER, streaming ML, and feature engineering.

Batch anomaly detection with 12+ backends. Train models, score data, detect outliers using isolation forests, autoencoders, and more.

Detect anomalies in data using batch or streaming ML models

REST API integration for LlamaFarm. Programmatic access to projects, chat, RAG, and datasets via HTTP endpoints.

Classify text using zero-shot or trained classification models

Generate LlamaFarm configurations from natural language. Activates for new projects, config creation, PDF/markdown/code patterns.

Schema-aware guidance for validating LlamaFarm configurations. Activates for llamafarm.yaml validation, config errors, schema reference.

Generate or modify LlamaFarm configuration from natural language description

Debugging and troubleshooting guide for LlamaFarm. Diagnose service issues, ingestion failures, retrieval problems, and performance bottlenecks.

Deployment guide for LlamaFarm. Production setup with Docker, Kubernetes, and cloud platforms.

Browse and scaffold from LlamaFarm example projects

Example projects and patterns for LlamaFarm. Reference for scaffolding and learning from working examples.

View LlamaFarm service logs with filtering

MCP (Model Context Protocol) development guide. Build custom tool servers, configure inline tools, and manage tool access in LlamaFarm.

OCR, Named Entity Recognition, and reranking. Extract text from images, identify entities in text, and rerank search results for relevance.

Check health and status of all ML services and models

End-to-end ML patterns and use cases. Complete pipelines for IoT monitoring, fraud detection, and document intelligence using LlamaFarm ML features.

Extract text from images and scanned documents using OCR

Polars-based feature engineering buffers. Create sliding windows, compute rolling statistics, and generate lag features for ML pipelines.

Deep guidance for RAG pipelines. Activates for chunking, embeddings, retrieval, vector stores, RAG optimization.

Start LlamaFarm services (server, RAG worker, optional Universal Runtime)

Check LlamaFarm service health and component status

Stop LlamaFarm services gracefully

Real-time streaming anomaly detection. Monitor data streams with automatic model updates, cold-start handling, and configurable retraining.

Zero-shot and custom text classification. Classify text with pre-trained models or train custom classifiers using SetFit with few-shot learning.

Validate llamafarm.yaml configuration against schema