ChatHistoryMemoryProvider Class

Definition

public sealed class ChatHistoryMemoryProvider : Microsoft.Agents.AI.MessageAIContextProvider, IDisposable

type ChatHistoryMemoryProvider = class
    inherit MessageAIContextProvider
    interface IDisposable

Public NotInheritable Class ChatHistoryMemoryProvider
Inherits MessageAIContextProvider
Implements IDisposable

Inheritance

Object

AIContextProvider

MessageAIContextProvider

ChatHistoryMemoryProvider

Implements

IDisposable

Remarks

This provider stores chat messages in a vector store and retrieves relevant previous messages to provide as context during agent invocations. It uses the VectorStore and VectorStoreCollection abstractions to work with any compatible vector store implementation.

Messages are stored during the Microsoft.Agents.AI.ChatHistoryMemoryProvider.StoreAIContextAsync(Microsoft.Agents.AI.AIContextProvider.InvokedContext,System.Threading.CancellationToken) method and retrieved during the Microsoft.Agents.AI.ChatHistoryMemoryProvider.ProvideAIContextAsync(Microsoft.Agents.AI.AIContextProvider.InvokingContext,System.Threading.CancellationToken) method using semantic similarity search.

Behavior is configurable through ChatHistoryMemoryProviderOptions. When OnDemandFunctionCalling is selected the provider exposes a function tool that the model can invoke to retrieve relevant memories on demand instead of injecting them automatically on each invocation.

Security considerations:

• Indirect prompt injection: Messages retrieved from the vector store via semantic search are injected into the LLM context. If the vector store is compromised, adversarial content could influence LLM behavior. The data returned from the store is accepted as-is without validation or sanitization.
• PII and sensitive data: Conversation messages (including user inputs and LLM responses) are stored as vectors in the underlying store. These messages may contain PII or sensitive information. Ensure the vector store is configured with appropriate access controls and encryption at rest.
• On-demand search tool: When using OnDemandFunctionCalling, the AI model controls when and what to search for. The search query is AI-generated and should be treated as untrusted input by the vector store implementation.
• Trace logging: When Trace is enabled, full search queries and results may be logged. This data may contain PII.