### Install Dev Environment and Run Example

Source: https://github.com/topoteretes/cognee/blob/main/examples/README.md

Installs the development environment, configures API keys by copying a template and editing it, and then runs a simple Cognee Python example.

```bash
# Install dev environment
uv sync --dev --all-extras --reinstall

# Configure API keys (one-time)
cp .env.template .env
# edit .env: set LLM_API_KEY (your OpenAI key) at minimum

# Run any example
uv run python examples/demos/simple_cognee_example.py
```

--------------------------------

### Quickstart MCP Server with Docker

Source: https://github.com/topoteretes/cognee/blob/main/AGENTS.md

Launch the MCP server using Docker for a quick local setup. This example uses HTTP transport and environment variables for configuration.

```bash
docker run -e TRANSPORT_MODE=http --env-file ./.env -p 8000:8000 --rm -it cognee/cognee-mcp:main
```

--------------------------------

### Install Frontend Dependencies and Start Dev Server

Source: https://github.com/topoteretes/cognee/blob/main/AGENTS.md

Commands to install Node.js dependencies and start the Next.js development server for the Cognee frontend. Also includes commands for linting and building.

```bash
cd cognee-frontend
npm install
npm run dev     # Next.js dev server
```

```bash
npm run lint    # ESLint
```

```bash
npm run build && npm start
```

--------------------------------

### Install Daytona Python SDK and Run Sandbox

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Install the Daytona Python SDK and set API keys. Then, run the provided Python script to start Cognee within a Daytona sandbox.

```bash
pip install daytona

export DAYTONA_API_KEY=your-key   # from https://app.daytona.io
export LLM_API_KEY=sk-xxx
python distributed/deploy/daytona_sandbox.py
```

--------------------------------

### Setup Cognee Development Environment

Source: https://github.com/topoteretes/cognee/blob/main/CLAUDE.md

Installs Cognee and its development dependencies using uv. Ensures pre-commit hooks are set up for code quality.

```bash
uv venv && source .venv/bin/activate

uv pip install -e .

uv pip install -e ".[dev]"

uv pip install -e ".[postgres,neo4j,docs,chromadb]"

pre-commit install
```

--------------------------------

### Install pre-commit and Install Hooks

Source: https://github.com/topoteretes/cognee/blob/main/CONTRIBUTING.md

Installs pre-commit and sets up the project's git hooks to ensure code style consistency.

```shell
uv run pip install pre-commit && pre-commit install
```

--------------------------------

### Install uv Package Manager

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Install uv, a fast Python package installer, if you don't have it already.

```bash
pip install uv
```

--------------------------------

### Start the development server

Source: https://github.com/topoteretes/cognee/blob/main/cognee-frontend/README.md

Use these commands to launch the local development environment.

```bash
npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev
```

--------------------------------

### Run Simple Cognee Example

Source: https://github.com/topoteretes/cognee/blob/main/CONTRIBUTING.md

Executes a simple example demonstrating Cognee functionality. Make sure to copy `.env.template` to `.env` and set your `OPENAI_API_KEY` as `LLM_API_KEY`. Also, run `uv sync` or set up a virtual environment.

```shell
uv run python examples/demos/simple_cognee_example.py
```

--------------------------------

### Create Daytona Sandbox and Install Cognee via CLI

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Install the Daytona CLI using Homebrew. Create a new sandbox, then install the Cognee API package and run the Uvicorn server within the sandbox.

```bash
brew install daytonaio/cli/daytona
daytona create
# Inside the sandbox:
pip install 'cognee[api]'
python -m uvicorn cognee.api.client:app --host 0.0.0.0 --port 8000
```

--------------------------------

### Start Cognee UI with Docker Compose

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Add the 'ui' profile to Docker Compose to start the frontend interface.

```bash
docker compose --profile ui up        # + frontend on http://localhost:3000
```

--------------------------------

### Start Cognee API Server with Docker Compose

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Use this command to start the main Cognee API server using Docker Compose.

```bash
docker compose up
```

--------------------------------

### Install Cognee using uv

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Install the Cognee library using the uv package manager. Ensure you have Python 3.10 to 3.14 installed.

```bash
uv pip install cognee
```

--------------------------------

### Execute Cognify Pipeline with Example Data

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/cognee_demo.ipynb

Retrieves a default user, fetches a specific dataset by name, and then initiates the cognify pipeline with the dataset and user. Ensure the 'example' dataset exists and user authentication is configured.

```python
from cognee.modules.users.methods import get_default_user
from cognee.modules.data.methods import get_datasets_by_name
from cognee.modules.users.methods import get_user

default_user = await get_default_user()

user = await get_user(default_user.id)

datasets = await get_datasets_by_name(["example"], user.id)

await run_cognify_pipeline(datasets[0], user)
```

--------------------------------

### Install Optional Dependencies from Docker Hub Image

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Runs a prebuilt cognee-mcp Docker image from Docker Hub, installing optional dependencies at runtime using the EXTRAS environment variable.

```bash
# Install optional dependencies from Docker Hub image
docker run \
  -e TRANSPORT_MODE=http \
  -e EXTRAS=aws,postgres \
  --env-file ./.env \
  -p 8000:8000 \
  --rm -it cognee/cognee-mcp:main
```

--------------------------------

### Install Cognee Tracing

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Install the necessary package for enabling OpenTelemetry tracing in Cognee.

```bash
pip install cognee[tracing]
```

--------------------------------

### Install and Run MCP Server Locally

Source: https://github.com/topoteretes/cognee/blob/main/AGENTS.md

Commands to install development dependencies and run the MCP server locally. Supports stdio, SSE, and HTTP transports with configurable hosts and ports.

```bash
cd cognee-mcp
uv sync --dev --all-extras --reinstall
uv run python src/server.py               # stdio (default)
```

```bash
uv run python src/server.py --transport sse
```

```bash
uv run python src/server.py --transport http --host 127.0.0.1 --port 8000 --path /mcp
```

--------------------------------

### Install Colima and Docker via Homebrew

Source: https://github.com/topoteretes/cognee/blob/main/docs/docker-colima-setup.md

Installs Colima and Docker using Homebrew on macOS or Linux.

```bash
# macOS (Homebrew)
brew install colima docker

# Linux (Homebrew)
brew install colima docker
```

--------------------------------

### Start Colima Container Runtime

Source: https://github.com/topoteretes/cognee/blob/main/docs/docker-colima-setup.md

Starts the Colima VM. The recommended `--network-address` flag provides a host-reachable network address for better host-container communication.

```bash
# Basic start
colima start

# Recommended: give the VM a host-reachable network address
colima start --network-address
```

--------------------------------

### Install Cognee Package

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/ontology_demo.ipynb

Installs the Cognee package. This is a prerequisite for using the framework.

```python
# Install required package
# !pip install cognee
```

--------------------------------

### Configure and perform text translation

Source: https://github.com/topoteretes/cognee/blob/main/cognee/tests/tasks/translation/README.md

Example showing how to set translation configuration and perform a direct text translation.

```python
import cognee
from cognee.tasks.translation import translate_text

# Configure translation (optional - defaults to LLM provider)
cognee.config.set_translation_config(
    provider="llm",           # Uses configured LLM (default)
    target_language="en",     # Target language code
    confidence_threshold=0.7  # Minimum confidence for language detection
)

# Translate text directly
result = await translate_text(
    text="Bonjour le monde",
    target_language="en"
)
print(result.translated_text)  # "Hello world"
```

--------------------------------

### Initialize and Deploy to Railway

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Initializes a new Railway project and deploys it. Assumes the Railway CLI is installed and configured.

```bash
railway init && railway up
```

--------------------------------

### Install uv dependency manager

Source: https://github.com/topoteretes/cognee/blob/main/cognee-starter-kit/README.md

Install the uv package manager to manage project dependencies.

```bash
pip install uv
```

--------------------------------

### Start Cognee MCP in Cloud Mode via CLI

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Use CLI flags to specify the Cognee Cloud or remote instance URL and API key when starting the server in Cloud Mode.

```bash
python src/server.py --serve-url https://your-instance.cognee.ai --serve-api-key ck_...
```

--------------------------------

### Start Cognee Visualization Server

Source: https://github.com/topoteretes/cognee/blob/main/cognee/skill.md

Launch a local server to interactively visualize the Cognee graph. Specify the port for the server.

```python
await cognee.start_visualization_server(port=8080)
```

--------------------------------

### Start Cognee MCP Server with Docker Compose

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Add the 'mcp' profile to Docker Compose to start the MCP server.

```bash
docker compose --profile mcp up       # + MCP server on http://localhost:8001
```

--------------------------------

### Install Development Dependencies (Python)

Source: https://github.com/topoteretes/cognee/blob/main/AGENTS.md

Use this command to create or refresh the Python development environment and install all necessary development dependencies and extras. Requires Python >= 3.10 and < 3.14.

```bash
uv sync --dev --all-extras --reinstall
```

--------------------------------

### Run Cognee with Docker Compose (Neo4j)

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Start Cognee using Docker Compose with Neo4j graph database integration.

```bash
# With Neo4j graph database
docker-compose --profile neo4j up
```

--------------------------------

### Copy and Edit Environment File

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Before starting Docker Compose, copy the template environment file and set your LLM_API_KEY.

```bash
cp .env.template .env   # then edit .env and set LLM_API_KEY
```

--------------------------------

### Run Cognee with Docker Compose (Postgres)

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Start Cognee using Docker Compose with PostgreSQL and pgvector support.

```bash
# With Postgres + pgvector
docker-compose --profile postgres up
```

--------------------------------

### Set alternative translation providers

Source: https://github.com/topoteretes/cognee/blob/main/cognee/tests/tasks/translation/README.md

Examples for switching the translation provider to Google Cloud or Azure.

```python
# Use Google Cloud Translate (requires GOOGLE_TRANSLATE_API_KEY)
cognee.config.set_translation_provider("google")

# Use Azure Translator (requires AZURE_TRANSLATOR_KEY and AZURE_TRANSLATOR_REGION)
cognee.config.set_translation_provider("azure")
```

--------------------------------

### Start Cognee with Postgres using Docker Compose

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Add the 'postgres' profile to Docker Compose to include a Postgres database with PGVector.

```bash
docker compose --profile postgres up  # + Postgres/PGVector
```

--------------------------------

### Install Cognee Memory Plugin for Claude Code

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Install the Cognee memory plugin for Claude Code from the marketplace. This is a one-time setup. Ensure environment variables are set before launching Claude Code.

```bash
claude plugin marketplace add topoteretes/cognee-integrations
claude plugin install cognee-memory@cognee
```

--------------------------------

### Minimal Environment Setup

Source: https://github.com/topoteretes/cognee/blob/main/CLAUDE.md

Configure essential environment variables for LLM API key and model. Defaults to OpenAI and local file-based databases if not specified.

```bash
# Minimal setup (defaults to OpenAI + local file-based databases)
LLM_API_KEY="your_openai_api_key"
LLM_MODEL="openai/gpt-5-mini"  # Default model
```

--------------------------------

### Run Cognee Platform Examples

Source: https://github.com/topoteretes/cognee/blob/main/CLAUDE.md

Demonstrates how to run Cognee using its Python SDK or the command-line interface (CLI). Includes options for adding data, cognifying, searching, deleting, and launching the UI.

```bash
uv run python examples/demos/simple_cognee_example.py

cognee-cli add "Your text here"
cognee-cli cognify
cognee-cli search "Your query"
cognee-cli delete --all

cognee-cli -ui
```

--------------------------------

### Reset Cognee System and Get Version

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/cognee_demo.ipynb

Resets the Cognee system by pruning data and metadata, then prints the installed Cognee version. Requires the 'cognee' library to be imported.

```python
# Reset the cognee system with the following command:

import cognee

await cognee.prune.prune_data()
await cognee.prune.prune_system(metadata=True)
print(cognee.__version__)

```

--------------------------------

### Structured Output Framework Configuration

Source: https://github.com/topoteretes/cognee/blob/main/CLAUDE.md

Choose the framework for structured output. Defaults to 'instructor' (via litellm). Alternatively, use 'baml' which requires the 'baml' extra and specific BAML configurations.

```bash
# Use Instructor (default, via litellm)
STRUCTURED_OUTPUT_FRAMEWORK="instructor"

# Or use BAML (requires baml extra: pip install cognee[baml])
STRUCTURED_OUTPUT_FRAMEWORK="baml"
BAML_LLM_PROVIDER=openai
BAML_LLM_MODEL="gpt-4o-mini"
BAML_LLM_API_KEY="your_api_key"
```

--------------------------------

### Install MCP Dependencies Locally

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Install and reinstall dependencies for the Cognee MCP project using uv. This step is required when using a local Cognee installation.

```bash
uv sync --reinstall
```

--------------------------------

### Verify Cognee Installation

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/tutorial.ipynb

Check the import path and installation status of the Cognee package.

```python
import cognee
import os
from pathlib import Path

print('🔍 Quick Cognee Import Check')
print('=' * 30)
print(f'📍 Cognee location: {cognee.__file__}')
print(f'📁 Package directory: {os.path.dirname(cognee.__file__)}')

# Check if it's local or installed
current_dir = Path.cwd()
cognee_path = Path(cognee.__file__)
if current_dir in cognee_path.parents:
    print('🏠 Status: LOCAL DEVELOPMENT VERSION')
else:
    print('📦 Status: INSTALLED PACKAGE')
```

--------------------------------

### Install Cognee TypeScript Package

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Use this command to install the @cognee/cognee-ts package for Node.js or browser environments.

```bash
npm install @cognee/cognee-ts
```

--------------------------------

### Entity Extraction System Example

Source: https://github.com/topoteretes/cognee/blob/main/cognee/infrastructure/llm/prompts/extract_entities_system.txt

Example response format for the entity extraction system. It includes entities with their name, type, and description.

```json
{
  "entities": [
    {
      "name": "Albert Einstein",
      "is_a": {
        "name": "PERSON",
        "description": "Entity type for person entities"
      },
      "description": "A theoretical physicist who developed the theory of relativity."
    },
    {
      "name": "Theory of Relativity",
      "is_a": {
        "name": "CONCEPT",
        "description": "Entity type for concept entities"
      },
      "description": "A physics theory describing the relationship between space and time."
    },
    {
      "name": "Princeton University",
      "is_a": {
        "name": "ORGANIZATION",
        "description": "Entity type for organization entities"
      },
      "description": "An Ivy League research university in Princeton, New Jersey."
    }
  ]
}
```

--------------------------------

### Run Cognee with Docker Compose (Minimal)

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Start Cognee using Docker Compose with a minimal configuration, including SQLite, LanceDB, and Ladybug, without external dependencies.

```bash
# Minimal (SQLite + LanceDB + Ladybug - no external deps)
docker-compose up cognee
```

--------------------------------

### Launch Graph Visualization Server via Python

Source: https://github.com/topoteretes/cognee/blob/main/CLAUDE.md

Starts the visualization server programmatically using Python. Specify the desired port for the server.

```python
from cognee.api.v1.visualize import start_visualization_server
await start_visualization_server(port=8080)
```

--------------------------------

### Start Cognee FastAPI Server

Source: https://github.com/topoteretes/cognee/blob/main/AGENTS.md

Directly start the FastAPI server for the Cognee API. This command is used to run the backend API service.

```bash
uv run python -m cognee.api.client
```

--------------------------------

### Start Cognee MCP in Cloud Mode via Environment Variables

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Configure Cloud Mode by setting COGNEE_SERVICE_URL and COGNEE_API_KEY environment variables before running the server. This provides a zero-configuration startup.

```bash
export COGNEE_SERVICE_URL="https://your-instance.cognee.ai"
export COGNEE_API_KEY="ck_..."
python src/server.py
```

--------------------------------

### Install Cognee with Postgres Support

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Install the Cognee package with the necessary dependencies for Postgres integration. This command ensures all required libraries for using Postgres as the memory backend are included.

```bash
pip install "cognee[postgres]"
```

--------------------------------

### Initialize Cognee and Add Multimedia Files

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/cognee_multimedia_demo.ipynb

Initializes Cognee by pruning existing data and system state, then adds specified multimedia files (MP3 and PNG) to be processed for knowledge graph creation.

```python
import cognee

# Create a clean slate for cognee -- reset data and system state
await cognee.prune.prune_data()
await cognee.prune.prune_system(metadata=True)

# Add multimedia files and make them available for cognify
await cognee.add([mp3_file_path, png_file_path])
```

--------------------------------

### Create Minimal Environment File

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Create a minimal .env file with your OpenAI API key for running Cognee containers.

```bash
echo 'LLM_API_KEY="YOUR_OPENAI_API_KEY"' > .env
```

--------------------------------

### Run Cognee with Docker Compose (Full Stack UI)

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Start the full Cognee stack, including the user interface, using Docker Compose.

```bash
# Full stack with UI
docker-compose --profile ui up
```

--------------------------------

### Run Cognee API Server Prebuilt Image

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Run the prebuilt Cognee API server image, mapping port 8000 and using the environment file.

```bash
docker run --env-file "./.env" -p 8000:8000 --rm -it cognee/cognee:main
```

--------------------------------

### Install Optional Dependencies at Runtime in Docker

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Runs the cognee-mcp Docker container with optional dependencies installed at runtime using the EXTRAS environment variable. Supports single or multiple comma-separated groups.

```bash
# Install a single optional dependency group at runtime
docker run \
  -e TRANSPORT_MODE=http \
  -e EXTRAS=aws \
  --env-file ./.env \
  -p 8000:8000 \
  --rm -it cognee/cognee-mcp:main
```

```bash
# Install multiple optional dependency groups at runtime (comma-separated)
docker run \
  -e TRANSPORT_MODE=sse \
  -e EXTRAS=aws,postgres,neo4j \
  --env-file ./.env \
  -p 8000:8000 \
  --rm -it cognee/cognee-mcp:main
```

--------------------------------

### Deploy Cognee to Modal

Source: https://github.com/topoteretes/cognee/blob/main/distributed/deploy/README.md

Installs the Modal CLI, sets the LLM API key, and deploys the FastAPI server. The endpoint URL will be available in the Modal dashboard.

```bash
pip install modal && modal setup

export LLM_API_KEY=sk-xxx
bash distributed/deploy/modal-deploy.sh
```

--------------------------------

### Start Cognee MCP Server with SSE Transport

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Use this command to start the Cognee MCP server with Server-Sent Events (SSE) transport enabled. Ensure you have Docker and the necessary .env file.

```bash
docker run -e TRANSPORT_MODE=sse --env-file ./.env -p 8000:8000 --rm -it cognee/cognee-mcp:main
```

--------------------------------

### Start Cognee MCP Server in API Mode

Source: https://github.com/topoteretes/cognee/blob/main/cognee-mcp/README.md

Initiate the Cognee MCP server in API Mode by providing the URL and authentication token of the running Cognee FastAPI server. Ensure the FastAPI server is started first.

```bash
# Start your Cognee FastAPI server first (default port 8000)
cd /path/to/cognee
python -m cognee.api.client

# Then start the MCP server in API mode
cd cognee-mcp
python src/server.py --api-url http://localhost:8000 --api-token YOUR_AUTH_TOKEN
```

--------------------------------

### Initialize Knowledge Graph

Source: https://github.com/topoteretes/cognee/blob/main/notebooks/cognee_multimedia_demo.ipynb

Use this command to trigger the cognify process for building the knowledge graph.

```python
await cognee.cognify()
```

--------------------------------

### Verify Docker Installation

Source: https://github.com/topoteretes/cognee/blob/main/docs/docker-colima-setup.md

Checks if the Docker daemon is running and accessible, and runs a test container.

```bash
docker info   # Should print server information without errors
docker run --rm hello-world
```

--------------------------------

### Deploy Cognee to Modal

Source: https://github.com/topoteretes/cognee/blob/main/README.md

Execute this script for serverless, auto-scaling, GPU workloads on Modal.

```bash
bash distributed/deploy/modal-deploy.sh
```