Brain Server - MCP Knowledge Embedding Service

What is Brain Server - MCP Knowledge Embedding Service?

Brain Server - MCP Knowledge Embedding Service is an MCP (Model Context Protocol) server for managing knowledge embeddings and vector search. It integrates with AI applications via the Model Context Protocol and organizes knowledge into domain-specific brains.

How to use Brain Server - MCP Knowledge Embedding Service?

Clone the repository, install dependencies (npm install), copy .env.example to .env and configure PORT, MONGODB_URI, EMBEDDING_MODEL, and MAX_CHUNK_SIZE. Build with npm run build, then run in development mode (npm run dev) or production mode (npm start). Use the exposed MCP tools such as addKnowledge, searchSimilar, updateKnowledge, deleteKnowledge, batchAddKnowledge, and getEmbedding.

Key features of Brain Server - MCP Knowledge Embedding Service

• Generate high-quality vector embeddings for knowledge content
• Perform semantic search based on meaning, not keywords
• Organize knowledge into domain-specific brains
• Retrieve surrounding context for better understanding
• Track progress of long-running operations in real time
• Full compliance with the Model Context Protocol

Use cases of Brain Server - MCP Knowledge Embedding Service

• Store and manage knowledge embeddings for AI assistant context
• Search large knowledge bases by semantic similarity
• Batch add multiple knowledge entries for bulk ingestion
• Organize knowledge into separate brains for different domains
• Monitor embedding and ingestion progress during processing

FAQ from Brain Server - MCP Knowledge Embedding Service

What is the default embedding model?

The default embedding model is Xenova/all-MiniLM-L6-v2, configured via the EMBEDDING_MODEL environment variable.

What database does the server require?

The server requires MongoDB, configured through the MONGODB_URI environment variable.

How do I configure the server?

Copy .env.example to .env and set PORT, MONGODB_URI, EMBEDDING_MODEL, and MAX_CHUNK_SIZE as needed.

How do I run the server in development mode?

Run npm run dev for development mode with hot reloading.

What MCP tools does the server expose?

The server exposes addKnowledge, searchSimilar, updateKnowledge, deleteKnowledge, batchAddKnowledge, and getEmbedding.