Quickstart

Install

The client ships inside LodeDB as the [cloud] extra:

pip install "lodedb[cloud]"

Sign in and mint a key

Signup is invite-gated during the beta. No account yet? Request access and we send a code.

lodedb cloud login
lodedb cloud tokens mint --kind secret --environment production \
  --scope write --scope read:search --scope read:text

login opens your browser for a one-time approval and stores the credential locally. It talks to the hosted control plane by default; pass --host only for a staging or self-hosted deployment.

The key is the whole setup: there is no store to create. It is bound to the environment, and any store name your code writes to provisions itself.

Put the printed key in your environment (self-hosted deployments also set ORECLOUD_HOST):

export ORECLOUD_TOKEN="ore_sk_..."

Store and recall

from lodedb.cloud import Client

client = Client()                   # org + environment come from the key

memory = client.store("user-42")    # one store per end user
memory.add("likes hiking near Seattle")

context = memory.context_block("plan my weekend")
print(context)

A few things happened implicitly:

• No create step per user. client.store(name) makes no HTTP call, and a store that doesn't exist yet is provisioned by its first write as its own LodeDB instance (server-side minilm embeddings, text exposed so context_block and get work). A new end user is just a new store name.
• The write is durable when add returns. Writes are accepted asynchronously (durable and ordered), then folded into the index within seconds.
• Read-your-writes. A search from this handle waits for this handle's own writes to become visible, so the context_block call already reflects the add above.

Auto-provisioning covers the memory workload. For different options up front (another embedding preset, bring-your-own vectors, text exposure off, encryption), register the store explicitly with lodedb cloud store create before its first write; see the SDK & CLI reference.

Search

search is classic top-k retrieval; recall takes a whole raw user message and lets the server derive sub-queries:

hits = memory.search("outdoor plans", k=5)
for score, id, metadata in hits:
    print(score, id, metadata)

hits = memory.recall("what should I do this weekend?")

Next steps