All Classes and Interfaces

Class
Description
Represents an attachment that can be sent to an LLM and displayed in messages.
An exception which is thrown when somebody attempts to use AI component features without activating the associated feature flag first.
Provides the AI components feature flag for AI-related features such as AIOrchestrator, modular upload components, and MessageListItem attachments.
Contributes tools and lifecycle hooks to an AIOrchestrator — domain-specific behaviour like populating a grid, building a chart, or filling a form from natural-language requests.
Interface for file upload components that are used in an AI conversation.
Interface for input components that are used in an AI conversation.
Represents a message in an AI conversation.
Interface for message list components that can display AI conversation messages.
Callback for attachment click events in the message list.
Orchestrator for AI-powered chat interfaces.
Builder for configuring and creating an AIOrchestrator instance.
Fluent API for restoring transient dependencies on an AIOrchestrator after deserialization.
Listener for attachment click events.
Event fired when an attachment in the message list is clicked.
Supported content type categories for attachments.
Represents a chat message in a conversation history.
The role of a message sender in a conversation.
How sure the LLM was about a value it produced.
Provider for database schema information and SQL query execution on behalf of an LLM.
Factory for creating reusable database LLMProvider.ToolSpec instances.
LangChain4j implementation of LLMProvider.
Framework-agnostic interface for Large Language Model (LLM) providers.
Represents a request to the LLM containing all necessary context, configuration, and tools.
A framework-agnostic tool definition that the LLM can invoke.
A rectangular area on a page of the source document.
A rectangle expressed as fractions of the surface it sits on, measured from the top-left corner.
Intercepts the user's input before the orchestrator acts on it.
Handle for completing a prompt postponed via RequestInterceptor.RequestInterceptEvent.postpone(Duration).
The content of one prompt, handed to a RequestInterceptor before the orchestrator acts on it.
Listener for request submission events.
Event fired just before the LLM stream opens for a prompt.
Listener for LLM response events.
Event fired when a turn ends, on success or failure.
Metadata about an LLM response, published by an LLMProvider through LLMProvider.LLMRequest.metadataSink() when the model reports it.
Token usage of a turn.
One snippet the LLM says it read to produce a value.
Where a SourceExtract sits inside its source document.
Spring AI implementation of LLMProvider.
Signals that a turn was ended because the model requested more tool calls than allowed.
Signals a tool failure whose message is safe to pass back to the LLM.
The source data the LLM reported alongside one value it produced: the snippets it read and how sure it was.