Set Up a RAG Pipeline

What is RAG?

Retrieval-Augmented Generation (RAG) extends a language model with an external knowledge base. Instead of relying solely on training data, the model retrieves relevant passages from your documents for every request and uses them as context — resulting in precise, source-grounded answers.

Typical use cases:

• Internal knowledge base or company FAQ
• Document assistant (contracts, manuals, reports)
• Support bot with up-to-date product information

Create a Knowledge Base

Navigate to Files & RAG in the left sidebar and click New Knowledge Base.

Give it a name and optionally a description, then click Create.

Upload Documents

Open the knowledge base and upload your content. You can:

• Upload individual files — select one or more files from your device
• Upload an entire folder — upload all files in a folder at once

After upload, Mycelis automatically indexes the documents: the content is split into chunks, embeddings are created, and the vectors are stored for semantic search. Depending on file size, indexing takes a few seconds to a couple of minutes.

Connect to an Agent

Once the knowledge base is indexed, connect it to an agent to start using it. See Configure an Agent for details on attaching a knowledge base via the Knowledge & Tools tab or the Workflow editor.

Next steps

• Configure an Agent — attach a knowledge base and configure routing
• MCP Tools — combine document knowledge with real-time external tools