Goto step

A pipeline step that enables dynamic flow control by jumping to specified nodes.

GoToStep(name, target, input_map=None, output_state=None, allow_end=True, retry_config=None, error_handler=None, cache=None)

Bases: BasePipelineStep, HasInputsMixin

A pipeline step that enables dynamic flow control by jumping to specified nodes.

This step allows for dynamic routing in pipelines by jumping to different nodes based on runtime conditions. The target can be a static node name, a state key, or a callable that determines the target at runtime.

Attributes:

Name	Type	Description
name	str	A unique identifier for the pipeline step.
target	str | Val | Callable[[dict[str, Any]], str]	The target node specification.
input_map	dict[str, str | Val]	Mapping of argument names to either a state/config key (str) or a fixed value (Val). Inherited from HasInputsMixin.
output_state	str | None	The state key to store the resolved target.
allow_end	bool	Whether jumping to 'END' is allowed.
retry_policy	RetryPolicy | None	Configuration for retry behavior using LangGraph's RetryPolicy. Inherited from BasePipelineStep.
cache_store	Any | None	The cache store used for caching step results, if configured. Inherited from BasePipelineStep.
is_cache_enabled	bool	Property indicating whether caching is enabled for this step. Inherited from BasePipelineStep.

Initializes a new GoToStep.

Parameters:

Name	Type	Description	Default
name	str	A unique identifier for this pipeline step.	required
target	str | Val | Callable[[dict[str, Any]], str]	The target node to jump to. Can be a literal string node name, a Val containing the node name, or a callable that takes the resolved inputs and returns the target node name.	required
input_map	InputMapSpec | None	Unified input map. Can be a dict (arg -> str|Val) or a list with elements: 1. str for identity mapping 2. dict[str, str] for state/config mapping 3. dict[str, Val] for fixed args. Defaults to None.	None
output_state	str | None	Key to store the target node name in the pipeline state. Defaults to None.	None
allow_end	bool	Whether to allow jumping to the END node. Defaults to True.	True
retry_config	RetryConfig | None	Configuration for retry behavior using GLLM Core's RetryConfig. Defaults to None, in which case no retry config is applied.	None
error_handler	BaseStepErrorHandler | None	Strategy to handle errors during execution. Defaults to None, in which case the RaiseStepErrorHandler is used.	None
cache	CacheConfig | None	Configuration for the cache store including the cache store instance and its settings. Defaults to None, in which case no caching will be used.	None

add_to_graph(graph, previous_endpoints, retry_policy=None)

Adds the GoToStep to the graph and creates edges.

Parameters:

Name	Type	Description	Default
graph	Any	The LangGraph graph to add the step to.	required
previous_endpoints	list[str]	List of source node names to connect from.	required
retry_policy	RetryPolicy | None	Optional retry policy to override the step's default. Defaults to None.	None

Returns:

Type	Description
list[str]	list[str]: List of endpoint names after adding this step.

apply_exclusions(exclusions)

Applies exclusions to this goto step.

Parameters:

Name	Type	Description	Default
exclusions	ExclusionSet	The exclusion set to apply.	required

execute(state, runtime, config=None) async

Executes the goto step and returns a LangGraph Command.

Parameters:

Name	Type	Description	Default
state	PipelineState	The current pipeline state.	required
runtime	Runtime	The runtime context containing config.	required
config	RunnableConfig | None	Optional runnable configuration. Defaults to None.	None

Returns:

Name	Type	Description
Command	Command	A LangGraph Command with goto parameter for routing.

Raises:

Type	Description
ValueError	If the resolved target is invalid.
KeyError	If state key is missing for string targets.

execute_direct(state, runtime, config=None) async

Executes the goto step and returns a state update.

Parameters:

Name	Type	Description	Default
state	PipelineState	The current pipeline state.	required
runtime	Runtime	The runtime context containing config.	required
config	RunnableConfig | None	Optional runnable configuration. Defaults to None.	None

Returns:

Type	Description
dict[str, Any] | None	dict[str, Any] | None: State update with resolved target if output_state is specified, None otherwise.

Raises:

Type	Description
ValueError	If the resolved target is invalid.
KeyError	If state key is missing for string targets.

get_mermaid_diagram(indent=0)

Generates a Mermaid diagram representation of this goto step.

Parameters:

Name	Type	Description	Default
indent	int	Number of spaces to indent the diagram. Defaults to 0.	0

Returns:

Name	Type	Description
str	str	The Mermaid diagram string.