The existing /v1/memories endpoint accepts a snapshot field instead of collection_id. When a snapshot is provided, Nebula processes the content statelessly and returns an updated snapshot instead of a memory_id.
POST /v1/memories
Field
Type
Required
Description
snapshot
object
Yes
Your current snapshot (mutually exclusive with collection_id)
raw_text
string
Yes
Content to store
metadata
object
No
Arbitrary metadata
Returns { "snapshot": <updated_snapshot> } when using device memory mode.
from nebula import Memoryresult = client.store_memory( Memory( snapshot=snapshot, content="Had a meeting with Alice about the Q2 roadmap", ))snapshot = result.snapshot
The existing /v1/memories/search endpoint accepts a snapshot field instead of collection_ids. Nebula scores your snapshot statelessly and returns results. Nothing is persisted.
POST /v1/memories/search
Field
Type
Required
Description
snapshot
object
Yes
Your full snapshot (mutually exclusive with collection_ids)
query
string
Yes
Natural language search query
Returns the same MemoryResponse format as standard search (semantics, procedures, episodes, sources).
results = client.search( query="What do I know about Alice?", snapshot=snapshot,)for fact in results.semantics: print(f"{fact['subject']} -> {fact['predicate']} -> {fact['value']}")