Vector Search

Vector Search enables AI Agents and AI Flows to retrieve relevant information from your workspace knowledge base using semantic search. Instead of relying on exact keyword matches, it understands the meaning and context of a query to return the most relevant results.

This skill is commonly used in Retrieval-Augmented Generation (RAG) workflows, allowing AI to answer questions using your organization’s documents and data.

Available Skills

Vector Search

Search embedded documents stored in the workspace knowledge base using semantic search.

The skill automatically searches indexed content and returns the most relevant document chunks, which can then be used by AI Agents to generate accurate, context-aware responses.

Common use cases

Question answering over internal documents
AI-powered knowledge assistants
Customer support automation
Document search
Semantic content retrieval
Retrieval-Augmented Generation (RAG)

How It Works

The Vector Search skill searches documents that have been indexed into your workspace’s knowledge base.

A typical retrieval process includes:

Upload documents to a Knowledge Base.
Split documents into searchable chunks.
Generate embeddings using a configured embedding model.
Store embeddings in a supported vector database.
Retrieve the most relevant chunks for a user query.
Provide the retrieved context to an AI Agent or AI Flow.

Typical Workflows

Vector Search integrates seamlessly with other Karyam capabilities.

Examples include:

Build an AI chatbot that answers questions from company documentation.
Search product manuals and generate accurate responses.
Retrieve relevant policies before responding to employee questions.
Combine semantic search with Web Search for richer AI responses.
Power AI Flows using enterprise knowledge stored in a Knowledge Base.