Context Grounding

ContextGroundingService ¶

Service for managing semantic automation contexts in UiPath.

Context Grounding is a feature that helps in understanding and managing the semantic context in which automation processes operate. It provides capabilities for indexing, retrieving, and searching through contextual information that can be used to enhance AI-enabled automation.

This service requires a valid folder key to be set in the environment, as context grounding operations are always performed within a specific folder context.

add_to_index ¶

add_to_index(
    name,
    blob_file_path,
    content_type=None,
    content=None,
    source_path=None,
    folder_key=None,
    folder_path=None,
    ingest_data=True,
)

Add content to the index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the index to add content to.	required
`content_type`	`str \| None`	The MIME type of the file. For file inputs this is computed dynamically. Default is "application/octet-stream".	`None`
`blob_file_path`	`str`	The path where the blob will be stored in the storage bucket.	required
`content`	`str \| bytes \| None`	The content to be added, either as a string or bytes.	`None`
`source_path`	`str \| None`	The source path of the content if it is being uploaded from a file.	`None`
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`
`ingest_data`	`bool`	Whether to ingest data in the index after content is uploaded. Defaults to True.	`True`

Raises:

Type	Description
`ValueError`	If neither content nor source_path is provided, or if both are provided.

add_to_index_async `async` ¶

add_to_index_async(
    name,
    blob_file_path,
    content_type=None,
    content=None,
    source_path=None,
    folder_key=None,
    folder_path=None,
    ingest_data=True,
)

Asynchronously add content to the index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the index to add content to.	required
`content_type`	`str \| None`	The MIME type of the file. For file inputs this is computed dynamically. Default is "application/octet-stream".	`None`
`blob_file_path`	`str`	The path where the blob will be stored in the storage bucket.	required
`content`	`str \| bytes \| None`	The content to be added, either as a string or bytes.	`None`
`source_path`	`str \| None`	The source path of the content if it is being uploaded from a file.	`None`
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`
`ingest_data`	`bool`	Whether to ingest data in the index after content is uploaded. Defaults to True.	`True`

Raises:

Type	Description
`ValueError`	If neither content nor source_path is provided, or if both are provided.

create_index ¶

create_index(
    name,
    source,
    description=None,
    advanced_ingestion=True,
    preprocessing_request=LLMV4_REQUEST,
    folder_key=None,
    folder_path=None,
)

Create a new context grounding index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the index to create.	required
`source`	`SourceConfig`	Source configuration using one of: - BucketSourceConfig: For storage buckets - GoogleDriveSourceConfig: For Google Drive - DropboxSourceConfig: For Dropbox - OneDriveSourceConfig: For OneDrive - ConfluenceSourceConfig: For Confluence The source can include an optional indexer field for scheduled indexing: source.indexer = Indexer(cron_expression="0 0 18 ? * 2", time_zone_id="UTC")	required
`description`	`str \| None`	Description of the index.	`None`
`advanced_ingestion`	`bool \| None`	Enable advanced ingestion with preprocessing. Defaults to True.	`True`
`preprocessing_request`	`str \| None`	The OData type for preprocessing request. Defaults to LLMV4_REQUEST.	`LLMV4_REQUEST`
`folder_key`	`str \| None`	The key of the folder where the index will be created.	`None`
`folder_path`	`str \| None`	The path of the folder where the index will be created.	`None`

Returns:

Name	Type	Description
`ContextGroundingIndex`	`ContextGroundingIndex`	The created index information.

create_index_async `async` ¶

create_index_async(
    name,
    source,
    description=None,
    advanced_ingestion=True,
    preprocessing_request=LLMV4_REQUEST,
    folder_key=None,
    folder_path=None,
)

Create a new context grounding index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the index to create.	required
`source`	`SourceConfig`	Source configuration using one of: - BucketSourceConfig: For storage buckets - GoogleDriveSourceConfig: For Google Drive - DropboxSourceConfig: For Dropbox - OneDriveSourceConfig: For OneDrive - ConfluenceSourceConfig: For Confluence The source can include an optional indexer field for scheduled indexing: source.indexer = Indexer(cron_expression="0 0 18 ? * 2", time_zone_id="UTC")	required
`description`	`str \| None`	Description of the index.	`None`
`advanced_ingestion`	`bool \| None`	Enable advanced ingestion with preprocessing. Defaults to True.	`True`
`preprocessing_request`	`str \| None`	The OData type for preprocessing request. Defaults to LLMV4_REQUEST.	`LLMV4_REQUEST`
`folder_key`	`str \| None`	The key of the folder where the index will be created.	`None`
`folder_path`	`str \| None`	The path of the folder where the index will be created.	`None`

Returns:

Name	Type	Description
`ContextGroundingIndex`	`ContextGroundingIndex`	The created index information.

delete_index ¶

delete_index(index, folder_key=None, folder_path=None)

Delete a context grounding index.

This method removes the specified context grounding index from Orchestrator.

Parameters:

Name	Type	Description	Default
`index`	`ContextGroundingIndex`	The context grounding index to delete.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

delete_index_async `async` ¶

delete_index_async(
    index, folder_key=None, folder_path=None
)

Asynchronously delete a context grounding index.

This method removes the specified context grounding index from Orchestrator.

Parameters:

Name	Type	Description	Default
`index`	`ContextGroundingIndex`	The context grounding index to delete.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

download_batch_transform_result ¶

download_batch_transform_result(
    id,
    destination_path,
    *,
    validate_status=True,
    index_name=None,
)

Downloads the Batch Transform result file to the specified path.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Batch Transform task.	required
`destination_path`	`str`	The local file path where the result file will be saved.	required
`validate_status`	`bool`	Whether to validate the batch transform status before downloading. Defaults to True.	`True`
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Raises:

Type	Description
`BatchTransformNotCompleteException`	If validate_status is True and the batch transform is not complete.

download_batch_transform_result_async `async` ¶

download_batch_transform_result_async(
    id,
    destination_path,
    *,
    validate_status=True,
    index_name=None,
)

Asynchronously downloads the Batch Transform result file to the specified path.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Batch Transform task.	required
`destination_path`	`str`	The local file path where the result file will be saved.	required
`validate_status`	`bool`	Whether to validate the batch transform status before downloading. Defaults to True.	`True`
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Raises:

Type	Description
`BatchTransformNotCompleteException`	If validate_status is True and the batch transform is not complete.

ingest_data ¶

ingest_data(index, folder_key=None, folder_path=None)

Ingest data into the context grounding index.

Parameters:

Name	Type	Description	Default
`index`	`ContextGroundingIndex`	The context grounding index to perform data ingestion.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

ingest_data_async `async` ¶

ingest_data_async(index, folder_key=None, folder_path=None)

Asynchronously ingest data into the context grounding index.

Parameters:

Name	Type	Description	Default
`index`	`ContextGroundingIndex`	The context grounding index to perform data ingestion.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

retrieve ¶

retrieve(name, folder_key=None, folder_path=None)

Retrieve context grounding index information by its name.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the context index to retrieve.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

Returns:

Name	Type	Description
`ContextGroundingIndex`	`ContextGroundingIndex`	The index information, including its configuration and metadata if found.

Raises:

Type	Description
`Exception`	If no index with the given name is found.

retrieve_async `async` ¶

retrieve_async(name, folder_key=None, folder_path=None)

Asynchronously retrieve context grounding index information by its name.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the context index to retrieve.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

Returns:

Name	Type	Description
`ContextGroundingIndex`	`ContextGroundingIndex`	The index information, including its configuration and metadata if found.

Raises:

Type	Description
`Exception`	If no index with the given name is found.

retrieve_batch_transform ¶

retrieve_batch_transform(id, *, index_name=None)

Retrieves a Batch Transform task status.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Batch Transform task.	required
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Returns:

Name	Type	Description
`BatchTransformResponse`	`BatchTransformResponse`	The Batch Transform task response.

retrieve_batch_transform_async `async` ¶

retrieve_batch_transform_async(id, *, index_name=None)

Asynchronously retrieves a Batch Transform task status.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Batch Transform task.	required
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Returns:

Name	Type	Description
`BatchTransformResponse`	`BatchTransformResponse`	The Batch Transform task response.

retrieve_by_id ¶

retrieve_by_id(id, folder_key=None, folder_path=None)

Retrieve context grounding index information by its ID.

This method provides direct access to a context index using its unique identifier, which can be more efficient than searching by name.

Parameters:

Name	Type	Description	Default
`id`	`str`	The unique identifier of the context index.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

Returns:

Name	Type	Description
`Any`	`Any`	The index information, including its configuration and metadata.

retrieve_by_id_async `async` ¶

retrieve_by_id_async(id, folder_key=None, folder_path=None)

Retrieve asynchronously context grounding index information by its ID.

This method provides direct access to a context index using its unique identifier, which can be more efficient than searching by name.

Parameters:

Name	Type	Description	Default
`id`	`str`	The unique identifier of the context index.	required
`folder_key`	`str \| None`	The key of the folder where the index resides.	`None`
`folder_path`	`str \| None`	The path of the folder where the index resides.	`None`

Returns:

Name	Type	Description
`Any`	`Any`	The index information, including its configuration and metadata.

retrieve_deep_rag ¶

retrieve_deep_rag(id, *, index_name=None)

Retrieves a Deep RAG task.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Deep RAG task.	required
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Returns:

Name	Type	Description
`DeepRagResponse`	`DeepRagResponse`	The Deep RAG task response.

retrieve_deep_rag_async `async` ¶

retrieve_deep_rag_async(id, *, index_name=None)

Asynchronously retrieves a Deep RAG task.

Parameters:

Name	Type	Description	Default
`id`	`str`	The id of the Deep RAG task.	required
`index_name`	`str \| None`	Index name hint for resource override.	`None`

Returns:

Name	Type	Description
`DeepRagResponse`	`DeepRagResponse`	The Deep RAG task response.

search ¶

search(
    name,
    query,
    number_of_results=10,
    folder_key=None,
    folder_path=None,
)

Search for contextual information within a specific index.

This method performs a semantic search against the specified context index, helping to find relevant information that can be used in automation processes. The search is powered by AI and understands natural language queries.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the context index to search in.	required
`query`	`str`	The search query in natural language.	required
`number_of_results`	`int`	Maximum number of results to return. Defaults to 10.	`10`

Returns:

Type	Description
`list[ContextGroundingQueryResponse]`	List[ContextGroundingQueryResponse]: A list of search results, each containing relevant contextual information and metadata.

search_async `async` ¶

search_async(
    name,
    query,
    number_of_results=10,
    folder_key=None,
    folder_path=None,
)

Search asynchronously for contextual information within a specific index.

This method performs a semantic search against the specified context index, helping to find relevant information that can be used in automation processes. The search is powered by AI and understands natural language queries.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the context index to search in.	required
`query`	`str`	The search query in natural language.	required
`number_of_results`	`int`	Maximum number of results to return. Defaults to 10.	`10`

Returns:

Type	Description
`list[ContextGroundingQueryResponse]`	List[ContextGroundingQueryResponse]: A list of search results, each containing relevant contextual information and metadata.

start_batch_transform ¶

start_batch_transform(
    name,
    index_name,
    prompt,
    output_columns,
    storage_bucket_folder_path_prefix=None,
    enable_web_search_grounding=False,
    folder_key=None,
    folder_path=None,
)

Starts a Batch Transform, task on the targeted index.

Batch Transform tasks are processing and transforming csv files from the index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the Deep RAG task.	required
`index_name`	`str`	The name of the context index to search in.	required
`prompt`	`str`	Describe the task: what to research, what to synthesize.	required
`output_columns`	`list[BatchTransformOutputColumn]`	The output columns to add into the csv.	required
`storage_bucket_folder_path_prefix`	`str`	The prefix pattern for filtering files in the storage bucket. Use "" to include all files. Defaults to "".	`None`
`enable_web_search_grounding`	`bool \| None`	Whether to enable web search. Defaults to False.	`False`
`folder_key`	`str`	The folder key where the index resides. Defaults to None.	`None`
`folder_path`	`str`	The folder path where the index resides. Defaults to None.	`None`

Returns:

Name	Type	Description
`BatchTransformCreationResponse`	`BatchTransformCreationResponse`	The batch transform task creation response.

start_batch_transform_async `async` ¶

start_batch_transform_async(
    name,
    index_name,
    prompt,
    output_columns,
    storage_bucket_folder_path_prefix=None,
    enable_web_search_grounding=False,
    folder_key=None,
    folder_path=None,
)

Asynchronously starts a Batch Transform, task on the targeted index.

Batch Transform tasks are processing and transforming csv files from the index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the Deep RAG task.	required
`index_name`	`str`	The name of the context index to search in.	required
`prompt`	`str`	Describe the task: what to research, what to synthesize.	required
`output_columns`	`list[BatchTransformOutputColumn]`	The output columns to add into the csv.	required
`storage_bucket_folder_path_prefix`	`str`	The prefix pattern for filtering files in the storage bucket. Use "" to include all files. Defaults to "".	`None`
`enable_web_search_grounding`	`bool \| None`	Whether to enable web search. Defaults to False.	`False`
`folder_key`	`str`	The folder key where the index resides. Defaults to None.	`None`
`folder_path`	`str`	The folder path where the index resides. Defaults to None.	`None`

Returns:

Name	Type	Description
`BatchTransformCreationResponse`	`BatchTransformCreationResponse`	The batch transform task creation response.

start_deep_rag ¶

start_deep_rag(
    name,
    index_name,
    prompt,
    glob_pattern="**",
    citation_mode=CitationMode.SKIP,
    folder_key=None,
    folder_path=None,
)

Starts a Deep RAG task on the targeted index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the Deep RAG task.	required
`index_name`	`str`	The name of the context index to search in.	required
`prompt`	`str`	Describe the task: what to research across documents, what to synthesize and how to cite sources.	required
`glob_pattern`	`str`	The glob pattern to search in the index. Defaults to "**".	`'**'`
`citation_mode`	`CitationMode`	The citation mode to use. Defaults to SKIP.	`SKIP`
`folder_key`	`str`	The folder key where the index resides. Defaults to None.	`None`
`folder_path`	`str`	The folder path where the index resides. Defaults to None.	`None`

Returns:

Name	Type	Description
`DeepRagCreationResponse`	`DeepRagCreationResponse`	The Deep RAG task creation response.

start_deep_rag_async `async` ¶

start_deep_rag_async(
    name,
    index_name,
    prompt,
    glob_pattern="**",
    citation_mode=CitationMode.SKIP,
    folder_key=None,
    folder_path=None,
)

Asynchronously starts a Deep RAG task on the targeted index.

Parameters:

Name	Type	Description	Default
`name`	`str`	The name of the Deep RAG task.	required
`index_name`	`str`	The name of the context index to search in.	required
`name`	`str`	The name of the Deep RAG task.	required
`prompt`	`str`	Describe the task: what to research across documents, what to synthesize and how to cite sources.	required
`glob_pattern`	`str`	The glob pattern to search in the index. Defaults to "**".	`'**'`
`citation_mode`	`CitationMode`	The citation mode to use. Defaults to SKIP.	`SKIP`
`folder_key`	`str`	The folder key where the index resides. Defaults to None.	`None`
`folder_path`	`str`	The folder path where the index resides. Defaults to None.	`None`

Returns:

Name	Type	Description
`DeepRagCreationResponse`	`DeepRagCreationResponse`	The Deep RAG task creation response.

Context Grounding