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,
    content_type,
    blob_file_path,
    content=None,
    source_path=None,
    folder_key=None,
    folder_path=None,
)

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`	The type of content being added.	required
`blob_file_path`	`str`	The path where the blob will be stored in the storage bucket.	required
`content`	`Optional[Union[str, bytes]]`	The content to be added, either as a string or bytes.	`None`
`source_path`	`Optional[str]`	The source path of the content if it is being uploaded from a file.	`None`
`folder_key`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	The path of the folder where the index resides.	`None`

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,
    content_type,
    blob_file_path,
    content=None,
    source_path=None,
    folder_key=None,
    folder_path=None,
)

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`	The type of content being added.	required
`blob_file_path`	`str`	The path where the blob will be stored in the storage bucket.	required
`content`	`Optional[Union[str, bytes]]`	The content to be added, either as a string or bytes.	`None`
`source_path`	`Optional[str]`	The source path of the content if it is being uploaded from a file.	`None`
`folder_key`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	The path of the folder where the index resides.	`None`

Raises:

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

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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	The path of the folder where the index resides.	`None`

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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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_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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	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`	`Optional[str]`	The key of the folder where the index resides.	`None`
`folder_path`	`Optional[str]`	The path of the folder where the index resides.	`None`

Returns:

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

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.

Context Grounding