kognita|Docs

Documentation

Kognita Developer Docs

Kognita is a managed knowledge-base API that lets you ingest text, generate vector embeddings, and query your data with full-text, semantic, or hybrid search, all via a single HTTP API.

Explore the docs

Quick start

Make your first API call in under 5 minutes.

Knowledge bases

Understand the core concept behind Kognita's data model.

Authentication

How to securely connect to Kognita's API and MCP endpoints

Search

Full-text, semantic, and hybrid search: when to use each.

API reference

Complete reference for every endpoint in the public API.

MCP server

Connect Claude, Cursor, or any MCP client to your knowledge bases.

How it works

  1. 1

    Create a knowledge base

    A named container with a vector embedding configuration (model + dimensions).

  2. 2

    Ingest content

    POST text to the knowledge base. Kognita chunks it, generates embeddings, and stores both.

  3. 3

    Search

    Query with full-text, semantic, or hybrid search. Get ranked results back as JSON.