Core Concepts

Add, Update & Delete

A memory moves through a small set of customer-visible operations: add it, inspect it, update its editable fields, summarize related records, or delete it.

Handle add outcomes

A successful add can create a memory or return a completed no-storage result. Branch on the response instead of assuming an ID exists.

Add and bulk outcomes

CREATED

Memory exists

Use the returned ID and object.

SKIPPED

Nothing stored

Read the reason and do not retry unchanged input.

REJECTED

Bulk item failed

Inspect and correct that item before retrying.

Bulk add accepts at most 50 items. Inspect every per-item outcome because one response can contain a mix.

Update editable fields

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
    project_id=os.environ["MEMORYSYNC_PROJECT_ID"],
    end_user_id="user-123",
)

updated = client.update(
    123,
    tags=["preference", "verified"],
    importance=0.8,
    metadata={"reviewed": True},
)

Delete memories you no longer need

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
    project_id=os.environ["MEMORYSYNC_PROJECT_ID"],
    end_user_id="user-123",
)

deleted_ids = client.forget([123], reason="user_request")

Use expiry for temporary context

Set ttl_minutes when the memory should stop applying after a known period. Use importance for weighting; do not use expiry as a ranking control.

Related operations

Relationships & Summaries

Connect records or condense a selected set.

Memory API

Review exact endpoint request and response schemas.

Was this page helpful?

← Previous

Memory Object

Retrieval & Filters