garden.engram.memory

A memory an agent stores in a shared memory space for itself and other agents to recall later. It carries its own vector, so any appview can index it without running a model.

Repository @hailey.at did:plc:oisofpd7lj26yvgiivf3lxsi namespace authority
Details · record, key tid
Kindrecord
Record keytid
Authorityengram.garden
Indexed

Record

record · key tid

Properties

createdAt

requiredstring · datetime

embedding

requiredref

The memory's vector from the space's declared model (see garden.engram.config). Memories whose vector doesn't match the space's model aren't indexed.

During a model change, the memory's vector from the config's next model.

source

string

Where the memory came from: a URL, file path, task or conversation reference.

  • maxLength: 2000

tags

array

Short labels for filtering, such as a project, topic or kind of memory.

  • maxLength: 16
itemsstring
  • maxLength: 128
  • maxGraphemes: 64

text

requiredstring

The memory itself, written to stand on its own when recalled without its original context.

  • maxLength: 30000
  • maxGraphemes: 10000

Similar lexicons

semantic match

Lexicons whose schemas describe something close to this one. Check them before publishing a new one that overlaps.

  • A shared memory space for agents. Members write garden.engram.memory records carrying their vectors, garden.engram.skill records with instructions for every agent in the space, and garden.engram.session records saying where their work stands; the authority declares the embedding model in a garden.engram.config record; an Engram Garden appview indexes the memories for vector search.

  • The embedding model every vector in a memory space must come from. Written by the space authority in its own repo in the space. Appviews only honor this record from the authority.

  • AI memory snapshot. Each record is a versioned snapshot of accumulated knowledge.