Quick Start

Run your own second brain locally with zero setup beyond Ollama.

Prerequisites

• Node.js 20+
• Ollama installed locally (or any OpenAI-compatible endpoint)

1. Install Docket

git clone https://github.com/yourusername/docket.git
cd docket && npm install

2. Start Ollama

ollama pull llama3.2
ollama pull nomic-embed-text
ollama serve

3. Start Docket

npm start

The server starts on http://localhost:3000.

4. Ingest your first memory

curl -X POST http://localhost:3000/ingest \
  -F "file=@tests/fixtures/bicycle.txt" \
  -F "async=false"

Response:

{
  "id": "mem_abc123",
  "status": "completed",
  "createdAt": "2026-05-09T12:00:00Z"
}

In rich mode you will also see sector, salience, and summary in the response.

5. Query your brain

curl -X POST http://localhost:3000/query \
  -H "Content-Type: application/json" \
  -d '{"question": "What did I capture about bicycles?"}'

Response:

{
  "answer": "You captured a note about a bicycle.",
  "sources": [
    { "memoryId": "mem_abc123", "score": 0.94 }
  ]
}

Run the verification script

./scripts/verify-hello-world.sh

Next steps

• Configure adapters
• Use the Config Builder
• Learn about memory modes
• Enable RBAC
• Set up hosting