Cognee UI

You can easily interact with cognee using its intuitive UI. Below you will find two ways to set it up and step by step tutorial over an example.

1. Prerequisites & (Optional) Environment Setup

Clone the cognee project or have it locally


git clone https://github.com/topoteretes/cognee.git

Navigate to cognee & install required dependencies:
```
cd cognee
poetry install
```
And for frontend:
```
cd cognee-fronted
npm install
```
Setup env variables (Optional)

Copy .env.template and rename it to .env and fill it based on your preferences.

You can skip this step and configure necessary variables in the UI.

2. Run cognee UI

You can run cognee UI in 2 different ways:

Run in Docker


cd cognee
docker compose --profile ui up

Backend and frontend seperately


cd cognee
python cognee/api/client.py


cd cognee-fronted
npm run dev

Now you can view the UI in http://localhost:3000/

Let’s walk through a simple example

Step 1 - Authorization

Cognee UI Authorization Page

Step 2 - Basic configuration

If you configure the environment variables in the env file above, cognee fills this part automatically. Otherwise you can configure here or make changes.

Cognee UI Basic Configuration Page

Step 3 - Add knowledge

When you click “Upload your data”, you will be prompted to select from your local files. You can select multiple files. In this example we will select one. When it’s completed, click “Next”.

Cognee UI Add Knowledge Page

Step 4 - Cognify!

This step automatically starts building a knowledge graph from the data you uploaded in the previous step using cognee’s default pipeline cognify. Depending on the size of your data, you might need to wait a few minutes but generally it is completed in seconds. When completed, click “Explore data”.

Cognee UI Cognify Page

Step 5 - Explore freshly generated knowledge graph

On the left side, you can visually explore your knowledge graph. Using the chat interface on the left side, you can query it with 3 different retrieval approaches.

Query insights from documents: uses searchType.INSIGHTS
Completion using Cognee’s graph based memory: uses searchType.GRAPH_COMPLETION
Completion using RAG: uses searchType.RAG_COMPLETION

Cognee UI Explore Page

Step 6 - Query your knowledge graph

Now you can experiement with the available search types. In this example, we’ll use Completion using Cognee’s graph based memory and ask questions about the data we added, the novel - Alice in Wonderland.

Cognee UI Query Page 1

Cognee UI Query Page 2

When you click X on the upper right corner, you’ll reach an overview page where you can see the dataset you populated with your files.

Step 7 - Dataset overview

Here you can add more files, view your files, or navigate to the Explore page again.

Cognee UI Database Overview Page

That’s it!

This is a quick tutorial to help you get started with cognee UI.

Join the Conversation!

Have questions? Join our community now to connect with professionals, share insights, and get your questions answered!