Memory Class
Protocol for conversation memory storage implementations.
Memory stores and retrieves conversation messages, enabling persistent context across multiple interactions with AI models.
Constructor
Memory(*args, **kwargs)Methods
| get_all |
Get all messages from memory. |
| push |
Add a message to memory. |
| set_all |
Replace all messages in memory with the provided list. |
get_all
Get all messages from memory.
async get_all() -> list[Union[microsoft.teams.ai.message.UserMessage, microsoft.teams.ai.message.ModelMessage, microsoft.teams.ai.message.SystemMessage, microsoft.teams.ai.message.FunctionMessage]]Returns
| Type | Description |
|---|---|
|
List of all stored messages in chronological order |
push
Add a message to memory.
async push(message: UserMessage | ModelMessage | SystemMessage | FunctionMessage) -> NoneParameters
| Name | Description |
|---|---|
|
message
Required
|
The message to store (user, model, system, or function message) |
set_all
Replace all messages in memory with the provided list.
async set_all(messages: list[Union[microsoft.teams.ai.message.UserMessage, microsoft.teams.ai.message.ModelMessage, microsoft.teams.ai.message.SystemMessage, microsoft.teams.ai.message.FunctionMessage]]) -> NoneParameters
| Name | Description |
|---|---|
|
messages
Required
|
New list of messages to store, replacing existing content |