ManifoldDocs
04/Knowledge (RAG)

Knowledge Projects

Turning documents and data into searchable knowledge + MCP servers.

A knowledge project ingests documents (PDF, text) and datasources, splits them into chunks, embeds them and stores them in a vector database (Milvus). The result is a searchable knowledge base and a generated MCP server.

  1. 1Upload documents or connect a datasource.
  2. 2Extracted text is split into chunks (Extracted Text tab).
  3. 3A RAG build runs; vectors are created.
  4. 4An MCP server (Docker image) is generated; you connect it to your tools.