@microsoft/teams-ai package

Classes

AssistantsPlanner	A Planner that uses the OpenAI Assistants API.
AI	AI System.
ActionAugmentationSection	A prompt section that renders a list of actions to the prompt.
ActionOutputMessage	A section capable of rendering user input text and images as a user message.
ActionPlanner	A planner that uses a Large Language Model (LLM) to generate plans.
ActionResponseValidator	Validates action calls returned by the model.
AdaptiveCards	AdaptiveCards class to enable fluent style registration of handlers related to Adaptive Cards.
Application	Application class for routing and processing incoming requests.
ApplicationBuilder	A builder class for simplifying the creation of an Application instance.
AssistantMessage	A message sent by the assistant.
AuthError	An error thrown when an authentication error occurs.
Authentication	User authentication service.
AuthenticationManager	The user authentication manager.
AzureContentSafetyModerator	An Azure OpenAI moderator that uses OpenAI's moderation API to review prompts and plans for safety.
ConversationHistory	A section that renders the conversation history.
DataSourceSection	A section that renders a data source to a prompt.
DefaultAugmentation	The default 'none' augmentation.
DefaultModerator	Default moderator created by the AI system if one isn't configured.
DefaultResponseValidator	Default response validator that always returns true.
FunctionCallMessage	An assistant message containing a function to call.
FunctionResponseMessage	Message containing the response to a function call.
GPT3Tokenizer	Tokenizer that uses GPT-3's encoder.
GPTTokenizer	Tokenizer that uses GPT's cl100k_base encoding. To use GPT-3 encoding, pass in an instance of GPT3Tokenizer.
GroupSection	A group of sections that will rendered as a single message.
JSONResponseValidator	Parses any JSON returned by the model and optionally verifies it against a JSON schema.
LLMClient	LLMClient class that's used to complete prompts.
LayoutEngine	Base layout engine that renders a set of auto, fixed, or proportional length sections.
LlamaModel	An AI model that can be used to complete prompts.
MemoryFork	Forks an existing memory.
MessageExtensions	MessageExtensions class to enable fluent style registration of handlers related to Message Extensions.
MonologueAugmentation	The 'monologue' augmentation.
OpenAIEmbeddings	A EmbeddingsModel for calling OpenAI and Azure OpenAI hosted models.
OpenAIModel	A PromptCompletionModel for calling OpenAI and Azure OpenAI hosted models.
OpenAIModerator	A moderator that uses OpenAI's moderation API to review prompts and plans for safety.
Prompt	Top level prompt section.
PromptManager	A filesystem based prompt manager.
PromptSectionBase	Abstract Base class for most prompt sections.
SequenceAugmentation	The 'sequence' augmentation.
StreamingResponse	A helper class for streaming responses to the client.
SystemMessage	A system message.
TaskModules	TaskModules class to enable fluent style registration of handlers related to Task Modules.
TeamsAdapter	An adapter that implements the Bot Framework Protocol and can be hosted in different cloud environments both public and private.
TeamsAttachmentDownloader	Downloads attachments from Teams using the bots access token.
TemplateSection	A template section that will be rendered as a message.
TestModel	A PromptCompletionModel used for testing.
TextDataSource	A data source that can be used to add a static block of text to a prompt.
TextSection	A section of text that will be rendered as a message.
ToolsAugmentation	The 'tools' augmentation is for enabling server-side action/tools calling. In the Teams AI Library, the equivalent to OpenAI's 'tools' functionality is called an 'action'. More information about OpenAI's tools can be found at OpenAI API docs. Therefore, tools/actions are defined in actions.json, and when 'tools' augmentation is set in config.json, the LLM model can specify which action(s) to call. To avoid using server-side tool-calling, do not set augmentation to 'tools' in config.json. Server-side tool-calling is not compatible with other augmentation types.
TurnState	Represents the turn state for a conversation. Turn state includes conversation state, user state, and temporary state. Provides methods to access, modify, and delete the state objects.
TurnStateEntry	Accessor class for managing an individual state scope.
UserInputMessage	A section capable of rendering user input text and images as a user message.
UserMessage	A user message.
Utilities	Utility functions for manipulating .

Interfaces

AssistantsPlannerOptions	Options for configuring the AssistantsPlanner.
AIOptions	Options for configuring the AI system.
ActionPlannerOptions	Options used to configure an ActionPlanner instance.
AdaptiveCard	Strongly typed Adaptive Card.
AdaptiveCardSearchResult	Individual result returned from AdaptiveCards.search() handler.
AdaptiveCardsOptions	Options for AdaptiveCards class.
AdaptiveCardsSearchParams	Parameters passed to AdaptiveCards.search() handler.
ApplicationOptions	Options for the Application class.
Augmentation	An augmentation is a component that can be added to a prompt template to add additional functionality to the prompt.
AugmentationConfig	Interface for the augmentation configuration portion of a prompt template.
AuthenticationOptions	The options to configure the authentication manager
AuthenticatorResult
AzureOpenAIEmbeddingsOptions	Options for configuring an OpenAIEmbeddings to generate embeddings using an Azure OpenAI hosted model.
AzureOpenAIModelOptions	Options for configuring an OpenAIModel to call an Azure OpenAI hosted model.
AzureOpenAIModeratorOptions	Options for the OpenAI based moderator.
BaseOpenAIEmbeddingsOptions	Base model options common to both OpenAI and Azure OpenAI services.
BaseOpenAIModelOptions	Base model options common to both OpenAI and Azure OpenAI services.
ChatCompletionAction	An action that can be called by an LLM.
Citation	Citations returned by the model.
ClientCitation
ConfiguredAIOptions	The configured options for the AI system after all defaults have been applied.
ConfiguredLLMClientOptions	The configuration of the LLMClient instance.
ConfiguredPromptManagerOptions	The configured PromptManager options.
DataSource	A data source that can be used to render text that's added to a prompt.
DefaultConversationState	Default conversation state
DefaultTempState	Default temp state
DefaultUserState	Default user state
EmbeddingsModel	An AI model that can be used to create embeddings.
EmbeddingsResponse	Response returned by a EmbeddingsClient.
FeedbackLoopData	Data returned when the thumbsup or thumbsdown button is clicked and response sent when enable_feedback_loop is set to true in the AI Module.
FunctionCall	A named function to call.
ImageContentPart
InnerMonologue	Structure used to track the inner monologue of an LLM.
InputFile	A file sent by the user to the bot.
InputFileDownloader	A plugin responsible for downloading files relative to the current user's input.
LLMClientOptions	Options for an LLMClient instance.
LlamaModelOptions
Memory	Represents a memory.
Message	A message object sent to or received from an LLM.
MessageContext
Moderator	A moderator is responsible for reviewing and approving AI prompts and plans.
OpenAIEmbeddingsOptions	Options for configuring an OpenAIEmbeddings to generate embeddings using an OpenAI hosted model.
OpenAILikeEmbeddingsOptions	Options for configuring an embeddings object that calls an OpenAI compliant endpoint.
OpenAIModelOptions	Options for configuring an OpenAIModel to call an OpenAI hosted model.
OpenAIModeratorOptions	Options for the OpenAI based moderator.
Plan	A plan is a set of commands that the AI system will execute.
Planner	A planner is responsible for generating a plan that the AI system will execute.
PredictedCommand	A predicted command is a command that the AI system should execute.
PredictedDoCommand	A predicted DO command is an action that the AI system should perform.
PredictedDoCommandAndHandler	Entities argument passed to the action handler for AI.DoCommandActionName.
PredictedSayCommand	A predicted SAY command is a response that the AI system should say.
PromptChunk	Streaming chunk passed in the sendChunk event of a PromptCompletionClient.
PromptCompletionModel	An AI model that can be used to complete prompts.
PromptCompletionModelEvents	Events emitted by a PromptCompletionModel.
PromptFunctions	A collection of functions that can be called from a prompt template string.
PromptManagerOptions	Options used to configure the prompt manager.
PromptResponseValidator	A validator that can be used to validate prompt responses.
PromptSection	A section that can be rendered to a prompt as either text or an array of Message objects.
PromptTemplate	Prompt template cached by the prompt manager.
PromptTemplateConfig	Serialized prompt template configuration.
Query	Query arguments for a search-based message extension.
RenderedPromptSection	The result of rendering a section.
SensitivityUsageInfo	Sensitivity usage info for content sent to the user. This is used to provide information about the content to the user.
TaskModulesOptions	Options for TaskModules class.
TeamsAttachmentDownloaderOptions	Options for the TeamsAttachmentDownloader class.
TextContentPart
Tokenizer	Copyright (c) Microsoft Corporation. All rights reserved. Licensed under the MIT License.
TooManyStepsParameters	Parameters passed to the AI.TooManyStepsActionName action.
TypeRecord
ValidatedChatCompletionAction	A validated action call.
Validation	Response returned by a PromptResponseValidator.

Type Aliases

ActionHandler	The code to execute when the action's name is triggered.
ActionPlannerPromptFactory	Factory function used to create a prompt template.
AuthErrorReason	Cause of an authentication error.
ConversationUpdateEvents	Conversation update events.
EEMethodReturnType
EmbeddingsResponseStatus	Status of the embeddings response.
EmitEventNames
EventNames
InnerEEMethodReturnType
ListenerType
MatchingKeys
MessageContentParts
MessageReactionEvents	Message reaction event types.
NoUndefined
OAuthSettings	Settings used to configure user authentication through the OAuthPrompt.
OnEventNames
OverriddenKeys
OverriddenMethods
PromptCompletionModelBeforeCompletionEvent	Type signature for the beforeCompletion event of a PromptCompletionModel.
PromptCompletionModelChunkReceivedEvent	Type signature for the chunkReceived event of a PromptCompletionModel.
PromptCompletionModelEmitter	Helper type that strongly types the the EventEmitter for a PromptCOmpletionModel instance.
PromptCompletionModelResponseReceivedEvent	Type signature for the responseReceived event of a PromptCompletionModel.
PromptFunction	A function that can be called from a prompt template string.
ReturnTypeOfMethod
ReturnTypeOfMethodIfExists
RouteHandler	Function for handling an incoming request.
RouteSelector	Function for selecting whether a route handler should be triggered.
Selector	A selector function for matching incoming activities.
SignInResponse	The sign in response.
SignInStatus	The sign in status.
StrictBroadcast
StrictEventEmitter
TeamsMessageEvents	Message related events.
TurnEvents	Turn event types.
VoidKeys

Enums

MessageExtensionsInvokeNames	Names of the invoke activities for Message Extensions.
ModerationSeverity	The moderation severity level.
TaskModuleInvokeNames