AIAgentGraphPipeline
Represents a pipeline for AI agent graph execution, extending the functionality of AIAgentPipeline. This class manages the execution of specific nodes in the pipeline using registered handlers.
Represents a pipeline for AI agent graph execution, extending the functionality of AIAgentPipeline. This class manages the execution of specific nodes in the pipeline using registered handlers.
Represents a pipeline for AI agent graph execution, extending the functionality of AIAgentPipeline. This class manages the execution of specific nodes in the pipeline using registered handlers.
Constructors
Properties
Provides access to a Clock instance representing the current system time. The Clock can be used to retrieve the current time, create date-time instances, or perform operations based on a specific moment in time.
Represents the configuration settings for the AI agent.
Functions
Collects metadata contributions from every feature that registered a handler via provideToolCallMetadata and returns the combined map.
Collects metadata contributions from every feature that registered a handler via provideToolCallMetadata and returns the combined map.
Collects metadata contributions from every feature that registered a handler via provideToolCallMetadata and returns the combined map.
Installs a feature into the pipeline with the provided configuration.
Installs a feature into the AI agent storage using the provided feature key, configuration, and implementation.
Installs a feature into the pipeline with the provided configuration.
Installs a feature into the pipeline with the provided configuration.
Intercepts and sets a handler to be invoked before an agent is closed.
Intercepts and sets a handler to be invoked before an agent is closed.
Intercepts the completion of an agent's operation and assigns a custom handler to process the result.
Intercepts the completion of an agent's operation and assigns a custom handler to process the result.
Intercepts and handles errors occurring during the execution of an AI agent's strategy.
Intercepts and handles errors occurring during the execution of an AI agent's strategy.
Intercepts on before an agent started to modify or enhance the agent.
Intercepts on before an agent started to modify or enhance the agent.
Intercepts environment creation to allow features to modify or enhance the agent environment.
Intercepts environment creation to allow features to modify or enhance the agent environment.
Intercepts LLM calls after they are made to process or log the response.
Intercepts LLM calls after they are made to process or log the response.
Intercepts errors during LLM calls.
Intercepts a failed call to the Language Learning Model (LLM) within a specific AI agent feature and delegates the handling of the failure context to the provided interceptor.
Intercepts LLM calls before they are made to modify or log the prompt.
Intercepts LLM calls before they are made to modify or log the prompt.
Intercepts streaming operations after they complete to perform post-processing or cleanup.
Intercepts streaming operations after they complete to perform post-processing or cleanup.
Intercepts errors during the streaming process.
Intercepts errors during the streaming process.
Intercepts stream frames as they are received during the streaming process.
Intercepts stream frames as they are received during the streaming process.
Intercepts streaming operations before they begin to modify or log the streaming request.
Intercepts streaming operations before they begin to modify or log the streaming request.
Intercepts node execution after it completes.
Intercepts node execution after it is successfully completed.
Intercepts and handles node execution errors for a given feature.
Intercepts node execution when it fails.
Intercepts node execution before it starts.
Intercepts node execution before it starts.
Sets up an interceptor to handle the completion of a strategy for the given feature.
Sets up an interceptor to handle the completion of a strategy for the given feature.
Intercepts the strategy starting event to perform actions when an agent strategy begins execution.
Intercepts starting strategy event to perform actions when an agent strategy begins execution.
Intercepts the completion of a subgraph execution and allows handling of the event.
Intercepts subgraph execution after it is successfully completed.
Intercepts and handles subgraph execution failures for a given feature.
Intercepts subgraph execution when it fails.
Intercepts the execution of a subgraph when it starts.
Intercepts subgraph execution before it starts.
Intercepts the result of a tool call with a custom handler for a specific feature.
Intercepts the result of a tool call with a custom handler for a specific feature.
Sets up an interception mechanism to handle tool call failures for a specific feature.
Sets up an interception mechanism to handle tool call failures for a specific feature.
Intercepts and handles tool calls for the specified feature.
Intercepts and handles tool calls for the specified feature.
Intercepts validation errors encountered during the execution of tools associated with the specified feature.
Intercepts validation errors encountered during the execution of tools associated with the specified feature.
Invoked before an agent is closed to perform necessary pre-closure operations.
Notifies all registered handlers that an agent has finished execution.
Transforms the agent environment by applying all registered environment transformers.
Notifies all registered handlers about an error that occurred during agent execution.
Notifies all registered handlers that an agent has started execution.
Notifies all registered LLM handlers after a language model call has completed.
Notifies all registered LLM handlers if a validation error occurs during a language model call.
Notifies all registered LLM handlers before a language model call is made.
Invoked after streaming from a language model completes.
Invoked if an error occurs during the streaming process.
Invoked when a stream frame is received during the streaming process.
Invoked before streaming from a language model begins.
Notifies all registered node handlers after a node has been executed.
Handles errors occurring during the execution of a node by invoking all registered node execution error handlers.
Notifies all registered node handlers before a node is executed.
Notifies all registered strategy handlers that a strategy has finished execution.
Notifies all registered strategy handlers that a strategy has started execution.
Notifies all registered subgraph handlers after a subgraph has been executed.
Notifies all registered subgraph handlers when a subgraph execution fails.
Notifies all registered subgraph handlers before a subgraph is executed.
Notifies all registered tool handlers about the result of a tool call.
Notifies all registered tool handlers when a tool call fails with an exception.
Notifies all registered tool handlers when a tool is called.
Notifies all registered tool handlers when a validation error occurs during a tool call.
Registers a handler that contributes per-call metadata before a tool executes.
Registers a handler that contributes metadata for every tool call.
Registers a handler that contributes metadata for every tool call.
Uninstalls the specified feature associated with the given key from storage.