Skip to main content
GET
/
api
/
v1
/
memories
Get memories
curl --request GET \
  --url https://api.evermind.ai/api/v1/memories \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "user_id": "user_123",
  "group_id": "group_456",
  "limit": 20,
  "offset": 0,
  "memory_type": "episodic_memory",
  "start_time": "2024-01-01T00:00:00",
  "end_time": "2024-12-31T23:59:59"
}
'
{
  "message": "Memory retrieval successful, retrieved 1 memories",
  "result": {
    "has_more": false,
    "memories": [
      {
        "content": "User discussed coffee during the project sync",
        "memory_type": "episodic_memory",
        "summary": "Project sync coffee note",
        "timestamp": "2024-01-15T10:30:00",
        "user_id": "user_123"
      }
    ],
    "metadata": {
      "memory_type": "fetch",
      "source": "fetch_mem_service",
      "user_id": "user_123"
    },
    "total_count": 100
  },
  "status": "ok"
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Memory fetch request

Used for GET /api/v1/memories endpoint.

Note:

  • user_id and group_id support special value MAGIC_ALL ("all") to skip filtering
  • Empty string or None for user_id/group_id filters for null/empty values
  • user_id and group_id cannot both be MAGIC_ALL
  • limit is capped at MAX_FETCH_LIMIT (500)
​
user_id
string | null

User ID

Example:

"user_123"

​
group_id
string | null

Group ID

Example:

"group_456"

​
limit
integer | null
default:40

Maximum number of memories to return

Required range: 1 <= x <= 500
Example:

20

​
offset
integer | null
default:0

Pagination offset

Required range: x >= 0
Example:

0

​
memory_type
enum<string> | null
default:episodic_memory

Memory type, enum values from MemoryType:

  • profile: user profile
  • episodic_memory: episodic memory (default)
  • foresight: prospective memory
  • event_log: event log (atomic facts)
Available options:
profile,
episodic_memory,
foresight,
event_log
Example:

"episodic_memory"

​
start_time
string | null

Start time for time range filtering (ISO 8601 format)

Example:

"2024-01-01T00:00:00"

​
end_time
string | null

End time for time range filtering (ISO 8601 format)

Example:

"2024-12-31T23:59:59"

Response

Successful Response

Memory fetch API response

Response for GET /api/v1/memories endpoint.

​
result
FetchMemResponse · object
required

Memory fetch result

​
status
string
default:ok

Response status

Examples:

"ok"

"failed"

​
message
string
default:""

Response message

Example:

"Operation successful"