Installation

This guide installs and runs Draft’n run locally with Docker services, Supabase, and the backend.

1) Start infrastructure services

From the repository root:

Terminal
bash Copy
cd services
docker compose up -d

This launches PostgreSQL, Redis, Qdrant, SeaweedFS, and (optionally) Prometheus/Tempo/Grafana.

2) Configure SeaweedFS S3

Create config/seaweedfs/s3_config.json from s3_config.json.example and set access keys. These will be reused in credentials.env.

3) Supabase (local)

From the repo root:

Terminal
bash Copy
supabase start
supabase status

Copy the anon and service role keys. Then run the Edge Functions runtime:

Terminal
bash Copy
supabase functions serve

4) Backend environment

Create credentials.env at the repo root (copy credentials.env.example) and fill variables per Configuration page. Minimal local values:

credentials.env
env Copy
ADA_DB_URL=postgresql://postgres:ada_password@localhost:5432/ada_backend
INGESTION_DB_URL=postgresql://postgres:ada_password@localhost:5432/ada_ingestion
TRACES_DB_URL=postgresql://postgres:ada_password@localhost:5432/ada_traces

QDRANT_CLUSTER_URL=http://localhost:6333
QDRANT_API_KEY=secret_api_key

S3_ENDPOINT_URL=http://localhost:8333
S3_ACCESS_KEY_ID=your_s3_access_key_id
S3_SECRET_ACCESS_KEY=your_s3_secret_access_key
S3_BUCKET_NAME=s3-backend
S3_REGION_NAME=us-east-1

SUPABASE_PROJECT_URL=http://localhost:54321
SUPABASE_PROJECT_KEY=your_anon_key
SUPABASE_SERVICE_ROLE_SECRET_KEY=your_service_role_key
SUPABASE_USERNAME=you@example.com
SUPABASE_PASSWORD=your_password
SUPABASE_BUCKET_NAME=ada-backend

ADA_URL=http://localhost:8000

Generate secrets:

Terminal
bash Copy
uv venv && source .venv/bin/activate
uv sync
uv run python -c "import secrets; print(secrets.token_hex(32))"   # BACKEND_SECRET_KEY
uv run python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"  # FERNET_KEY
uv run python -c "from ada_backend.services.api_key_service import _generate_api_key, _hash_key; key=_generate_api_key(); print('INGESTION_API_KEY=',key); print('INGESTION_API_KEY_HASHED=',_hash_key(key))"

5) Initialize databases

Terminal
bash Copy
make db-upgrade
make db-seed
make trace-db-upgrade

6) Run backend and worker

Terminal
bash Copy
make run-draftnrun-agents-backend
# in another terminal
uv run python -m ada_ingestion_system.worker.main

Open:

Swagger: http://localhost:8000/docs
Admin: http://localhost:8000/admin
Grafana (if enabled): http://localhost:3000

Quick Start tutorial

Ready to build production AI?

Start with the Quick Start guide or explore the API. Join the community and help shape the open-source platform for AI agents.