Skip to main content
POST
/
api
/
v0
/
memories
/
conversation-meta
Python
# To use the Python SDK, install the package:
# pip install evermemos

from evermemos import EverMemOS

meta = EverMemOS(api_key="<api_key>").v0.memories.conversation_meta

response = meta.create(
    scene="<scene>",
    created_at="<iso8601_time>"
)
print(response.status)
{
  "message": "Conversation metadata saved successfully",
  "result": {
    "conversation_created_at": "2025-01-15T10:00:00+00:00",
    "group_id": "group_123",
    "id": "507f1f77bcf86cd799439011",
    "is_default": false,
    "name": "Project Discussion",
    "scene": "group_chat",
    "scene_desc": {
      "description": "Project discussion group chat"
    }
  },
  "status": "ok"
}

Authorizations

​
Authorization
string
header
default:Bearer <api_key>
required

Bearer authentication header of the form Bearer 'api_key', where 'api_key' is your EverMemOS auth api key.

Body

application/json
​
scene
string | null
required

Scene identifier.

Enum values from ScenarioType:

  • assistant: Ideal for personal assistants, AI companion or multi-party chats centered on the primary user.
  • group_chat: Tailored for group syncs and professional teamwork. It extracts independent, granular memories for every participant.
Example:

"assistant"

​
created_at
string
required

Conversation creation time (ISO 8601 format with Timezone is required)

Example:

"2025-01-15T10:00:00+00:00"

​
scene_desc
Scene Desc · object

Scene description object.

Can include fields like description, type, etc.

Example:
{
  "description": "Project discussion group chat",
  "type": "project_discussion"
}
​
llm_custom_setting
LlmCustomSetting · object

LLM custom settings for algorithm control.

Allows configuring different LLM providers/models for different tasks like boundary detection and memory extraction.

Example:
{
  "boundary": {
    "model": "qwen/qwen3-235b-a22b-2507",
    "provider": "openrouter"
  },
  "extraction": {
    "model": "qwen/qwen3-235b-a22b-2507",
    "provider": "openrouter"
  }
}
​
description
string | null

Conversation description

Example:

"Technical discussion for new feature development"

​
default_timezone
string | null

Default timezone

Example:

"UTC"

​
user_details
User Details · object

Participant details, key is user ID, value is user detail object

Example:
{
  "bot_001": {
    "custom_role": "assistant",
    "extra": { "type": "ai" },
    "full_name": "AI Assistant",
    "role": "assistant"
  },
  "user_001": {
    "custom_role": "developer",
    "extra": { "department": "Engineering" },
    "full_name": "John Smith",
    "role": "user"
  }
}
​
tags
string[] | null

Tag list

Example:
["work", "technical"]

Response

Successful Response

​
result
ConversationMetaResponse · object
required

Saved conversation metadata

Example:
{
  "conversation_created_at": "2025-01-15T10:00:00+00:00",
  "created_at": "2025-01-15T10:00:00+00:00",
  "default_timezone": "UTC",
  "description": "Technical discussion group",
  "id": "507f1f77bcf86cd799439011",
  "is_default": false,
  "scene": "assistant",
  "scene_desc": {
    "description": "Project discussion group chat"
  },
  "tags": ["work", "tech"],
  "updated_at": "2025-01-15T10:00:00+00:00",
  "user_details": {
    "bot_001": {
      "full_name": "AI Assistant",
      "role": "assistant"
    },
    "user_001": {
      "custom_role": "developer",
      "full_name": "John",
      "role": "user"
    }
  }
}
​
status
string
default:ok

Response status

Examples:

"ok"

"failed"

​
message
string
default:""

Response message

Example:

"Operation successful"