Class AbstractStreamParsingChatAgent`Abstract`

A chat agent is a specialized agent with a common interface for its invocation.

Hierarchy (view full)

AbstractChatAgent
- AbstractStreamParsingChatAgent
  - ChatSessionSummaryAgent
  - CustomChatAgent

Constructors

constructor

new AbstractStreamParsingChatAgent(): AbstractStreamParsingChatAgent
Returns AbstractStreamParsingChatAgent
Inherited from AbstractChatAgent.constructor

Properties

`Protected` additionalToolRequests

additionalToolRequests: ToolRequest[] = []

agentSpecificVariables

agentSpecificVariables: AgentSpecificVariables[] = []

`Protected` chatToolRequestService

chatToolRequestService: ChatToolRequestService

`Protected` contentMatcherProviders

contentMatcherProviders: ContributionProvider<ResponseContentMatcherProvider>

`Protected` contentMatchers

contentMatchers: ResponseContentMatcher[] = []

`Protected` defaultContentFactory

defaultContentFactory: DefaultResponseContentFactory

`Protected` `Readonly` `Abstract` defaultLanguageModelPurpose

defaultLanguageModelPurpose: string

description

description: string = ''

functions

functions: string[] = []

iconClass

iconClass: string = 'codicon codicon-copilot'

`Readonly` `Abstract` id

id: string

`Protected` languageModelRegistry

languageModelRegistry: LanguageModelRegistry

`Readonly` `Abstract` languageModelRequirements

languageModelRequirements: LanguageModelRequirement[]

`Protected` languageModelService

languageModelService: LanguageModelService

locations

locations: ChatAgentLocation[] = ChatAgentLocation.ALL

`Protected` logger

logger: ILogger

`Readonly` `Abstract` name

name: string

`Protected` promptService

promptService: PromptService

prompts

prompts: PromptVariantSet[] = []

`Protected` systemPromptId

systemPromptId: undefined | string = undefined

`Protected` toolCallResponseContentFactory

toolCallResponseContentFactory: ToolCallChatResponseContentFactory

variables

variables: string[] = []

Methods

`Protected` addContentsToResponse

addContentsToResponse(languageModelResponse, request): Promise<void>
Parameters
- languageModelResponse: LanguageModelResponse
- request: MutableChatRequestModel
Returns Promise<void>
Overrides AbstractChatAgent.addContentsToResponse
- Defined in packages/ai-chat/src/common/chat-agents.ts:359

`Protected` addStreamResponse

addStreamResponse(languageModelResponse, request): Promise<void>
Parameters
- languageModelResponse: LanguageModelStreamResponse
- request: MutableChatRequestModel
Returns Promise<void>
- Defined in packages/ai-chat/src/common/chat-agents.ts:379

`Protected` createToolCallResponseContent

createToolCallResponseContent(toolCall): ChatResponseContent
Creates a ToolCallChatResponseContent instance from the provided tool call data.

This method is called when parsing stream response tokens that contain tool call data. Subclasses can override this method to customize the creation of tool call response contents.
Parameters
- toolCall: ToolCall
  The ToolCall.
Returns ChatResponseContent
A ChatResponseContent representing the tool call.
- Defined in packages/ai-chat/src/common/chat-agents.ts:452

`Protected` getLanguageModel

getLanguageModel(languageModelPurpose): Promise<LanguageModel>
Parameters
- languageModelPurpose: string
Returns Promise<LanguageModel>
Inherited from AbstractChatAgent.getLanguageModel
- Defined in packages/ai-chat/src/common/chat-agents.ts:229

`Protected` getLanguageModelSelector

getLanguageModelSelector(languageModelPurpose): LanguageModelRequirement
Parameters
- languageModelPurpose: string
Returns LanguageModelRequirement
Inherited from AbstractChatAgent.getLanguageModelSelector
- Defined in packages/ai-chat/src/common/chat-agents.ts:225

`Protected` getLlmSettings

getLlmSettings(): undefined | {
[key: string]: unknown;
}
Returns undefined | {
[key: string]: unknown;
}
the settings, such as temperature, to be used in all language model requests. Returns undefined by default.
Inherited from AbstractChatAgent.getLlmSettings
- Defined in packages/ai-chat/src/common/chat-agents.ts:306

`Protected` getMessages

getMessages(model, includeResponseInProgress?): Promise<LanguageModelMessage[]>
Parameters
- model: ChatModel
- includeResponseInProgress: boolean = false
Returns Promise<LanguageModelMessage[]>
Inherited from AbstractChatAgent.getMessages
- Defined in packages/ai-chat/src/common/chat-agents.ts:249

`Protected` getSystemMessageDescription

getSystemMessageDescription(context): Promise<undefined | SystemMessageDescription>
Parameters
- context: AIVariableContext
Returns Promise<undefined | SystemMessageDescription>
Inherited from AbstractChatAgent.getSystemMessageDescription
- Defined in packages/ai-chat/src/common/chat-agents.ts:241

`Protected` handleError

handleError(request, error): void
Parameters
- request: MutableChatRequestModel
- error: Error
Returns void
Inherited from AbstractChatAgent.handleError
- Defined in packages/ai-chat/src/common/chat-agents.ts:220

init

init(): void
Returns void
Inherited from AbstractChatAgent.init
- Defined in packages/ai-chat/src/common/chat-agents.ts:167

`Protected` initializeContentMatchers

initializeContentMatchers(): void
Returns void
Inherited from AbstractChatAgent.initializeContentMatchers
- Defined in packages/ai-chat/src/common/chat-agents.ts:171

invoke

invoke(request): Promise<void>
Parameters
- request: MutableChatRequestModel
Returns Promise<void>
Inherited from AbstractChatAgent.invoke
- Defined in packages/ai-chat/src/common/chat-agents.ts:176

`Protected` onResponseComplete

onResponseComplete(request): Promise<void>
Invoked after the response by the LLM completed successfully.

The default implementation sets the state of the response to complete. Subclasses may override this method to perform additional actions or keep the response open for processing further requests.
Parameters
- request: MutableChatRequestModel
Returns Promise<void>
Inherited from AbstractChatAgent.onResponseComplete
- Defined in packages/ai-chat/src/common/chat-agents.ts:316

`Protected` parse

parse(token, request): ChatResponseContent | ChatResponseContent[]
Parameters
- token: LanguageModelStreamResponsePart
- request: MutableChatRequestModel
Returns ChatResponseContent | ChatResponseContent[]
- Defined in packages/ai-chat/src/common/chat-agents.ts:417

`Protected` parseContents

parseContents(text, request): ChatResponseContent[]
Parameters
- text: string
- request: MutableChatRequestModel
Returns ChatResponseContent[]
Inherited from AbstractChatAgent.parseContents
- Defined in packages/ai-chat/src/common/chat-agents.ts:211

`Protected` selectLanguageModel

selectLanguageModel(selector): Promise<LanguageModel>
Parameters
- selector: LanguageModelRequirement
Returns Promise<LanguageModel>
Inherited from AbstractChatAgent.selectLanguageModel
- Defined in packages/ai-chat/src/common/chat-agents.ts:233

`Protected` sendLlmRequest

sendLlmRequest(request, messages, toolRequests, languageModel): Promise<LanguageModelResponse>
Parameters
- request: MutableChatRequestModel
- messages: LanguageModelMessage[]
- toolRequests: ChatToolRequest[]
- languageModel: LanguageModel
Returns Promise<LanguageModelResponse>
Inherited from AbstractChatAgent.sendLlmRequest
- Defined in packages/ai-chat/src/common/chat-agents.ts:280

Class AbstractStreamParsingChatAgentAbstract

Hierarchy (view full)

Index

Constructors

Properties

Methods

Constructors

constructor

Returns AbstractStreamParsingChatAgent

Properties

Protected additionalToolRequests

agentSpecificVariables

Protected chatToolRequestService

Protected contentMatcherProviders

Protected contentMatchers

Protected defaultContentFactory

Protected Readonly Abstract defaultLanguageModelPurpose

description

functions

iconClass

Readonly Abstract id

Protected languageModelRegistry

Readonly Abstract languageModelRequirements

Protected languageModelService

locations

Protected logger

Readonly Abstract name

Protected promptService

prompts

Protected systemPromptId

tags

Protected toolCallResponseContentFactory

variables

Methods

Protected addContentsToResponse

Parameters

Returns Promise<void>

Protected addStreamResponse

Parameters

Returns Promise<void>

Protected createToolCallResponseContent

Parameters

Returns ChatResponseContent

Protected getLanguageModel

Parameters

Returns Promise<LanguageModel>

Protected getLanguageModelSelector

Parameters

Returns LanguageModelRequirement

Protected getLlmSettings

Returns undefined | { [key: string]: unknown; }

Protected getMessages

Parameters

Returns Promise<LanguageModelMessage[]>

Protected getSystemMessageDescription

Parameters

Returns Promise<undefined | SystemMessageDescription>

Protected handleError

Parameters

Returns void

init

Returns void

Protected initializeContentMatchers

Returns void

invoke

Parameters

Returns Promise<void>

Protected onResponseComplete

Parameters

Returns Promise<void>

Protected parse

Parameters

Returns ChatResponseContent | ChatResponseContent[]

Protected parseContents

Parameters

Returns ChatResponseContent[]

Protected selectLanguageModel

Parameters

Returns Promise<LanguageModel>

Protected sendLlmRequest

Class AbstractStreamParsingChatAgent`Abstract`

`Protected` additionalToolRequests

`Protected` chatToolRequestService

`Protected` contentMatcherProviders

`Protected` contentMatchers

`Protected` defaultContentFactory

`Protected` `Readonly` `Abstract` defaultLanguageModelPurpose

`Readonly` `Abstract` id

`Protected` languageModelRegistry

`Readonly` `Abstract` languageModelRequirements

`Protected` languageModelService

`Protected` logger

`Readonly` `Abstract` name

`Protected` promptService

`Protected` systemPromptId

`Protected` toolCallResponseContentFactory

`Protected` addContentsToResponse

`Protected` addStreamResponse

`Protected` createToolCallResponseContent

`Protected` getLanguageModel

`Protected` getLanguageModelSelector

`Protected` getLlmSettings

Returns undefined | {
[key: string]: unknown;
}

`Protected` getMessages

`Protected` getSystemMessageDescription

`Protected` handleError

`Protected` initializeContentMatchers

`Protected` onResponseComplete

`Protected` parse

`Protected` parseContents

`Protected` selectLanguageModel

`Protected` sendLlmRequest