Skip to main content

AI Integrations

WSO2 Integrator lets you build AI-powered integrations, including direct LLM calls, RAG pipelines, AI agents, and MCP servers.

Getting started​

Develop AI applications​

  • Direct LLM calls: The simplest AI block. Send a prompt and bind the response to a typed value, in a single round-trip.
  • RAG: Retrieval-augmented generation that grounds LLM responses in your own documents by retrieving relevant content at query time and injecting it into the prompt.
  • AI agents: Autonomous LLM-driven agents that reason over a system prompt, call tools, and maintain conversation state across turns.
  • MCP integration: Expose your integrations as MCP tools for AI assistants, or use external MCP tools with your agents.
  • Natural functions: (Experimental) Write the function body in plain English. The LLM returns a value that conforms to your declared return type.
  • Model providers: Connect to OpenAI, Anthropic, Azure OpenAI, Google Vertex, Mistral, and others through one consistent interface.
  • Embedding providers: Turn text into semantic vectors used on both ingest and query for similarity search.
  • Vector stores: Persist embeddings and run similarity search across in-memory, Pinecone, pgvector, Weaviate, or Milvus vector databases.
  • Knowledge bases: The indexable document store RAG reads from and writes to, composed of a vector store, embedding provider, and chunker.
  • Chunkers: Split documents into chunks before embedding. Smaller chunks improve retrieval precision. Larger chunks preserve more surrounding context.

Tutorials​