QConnect / Client / create_knowledge_base

create_knowledge_base#

QConnect.Client.create_knowledge_base(**kwargs)#

Creates a knowledge base.

Note

When using this API, you cannot reuse Amazon AppIntegrations DataIntegrations with external knowledge bases such as Salesforce and ServiceNow. If you do, you’ll get an InvalidRequestException error.

For example, you’re programmatically managing your external knowledge base, and you want to add or remove one of the fields that is being ingested from Salesforce. Do the following:

Call DeleteKnowledgeBase.
Call DeleteDataIntegration.
Call CreateDataIntegration to recreate the DataIntegration or a create different one.
Call CreateKnowledgeBase.

Request Syntax

response = client.create_knowledge_base(
    clientToken='string',
    description='string',
    knowledgeBaseType='EXTERNAL'|'CUSTOM'|'QUICK_RESPONSES'|'MESSAGE_TEMPLATES'|'MANAGED',
    name='string',
    renderingConfiguration={
        'templateUri': 'string'
    },
    serverSideEncryptionConfiguration={
        'kmsKeyId': 'string'
    },
    sourceConfiguration={
        'appIntegrations': {
            'appIntegrationArn': 'string',
            'objectFields': [
                'string',
            ]
        },
        'managedSourceConfiguration': {
            'webCrawlerConfiguration': {
                'crawlerLimits': {
                    'rateLimit': 123
                },
                'exclusionFilters': [
                    'string',
                ],
                'inclusionFilters': [
                    'string',
                ],
                'scope': 'HOST_ONLY'|'SUBDOMAINS',
                'urlConfiguration': {
                    'seedUrls': [
                        {
                            'url': 'string'
                        },
                    ]
                }
            }
        }
    },
    tags={
        'string': 'string'
    },
    vectorIngestionConfiguration={
        'chunkingConfiguration': {
            'chunkingStrategy': 'FIXED_SIZE'|'NONE'|'HIERARCHICAL'|'SEMANTIC',
            'fixedSizeChunkingConfiguration': {
                'maxTokens': 123,
                'overlapPercentage': 123
            },
            'hierarchicalChunkingConfiguration': {
                'levelConfigurations': [
                    {
                        'maxTokens': 123
                    },
                ],
                'overlapTokens': 123
            },
            'semanticChunkingConfiguration': {
                'breakpointPercentileThreshold': 123,
                'bufferSize': 123,
                'maxTokens': 123
            }
        },
        'parsingConfiguration': {
            'bedrockFoundationModelConfiguration': {
                'modelArn': 'string',
                'parsingPrompt': {
                    'parsingPromptText': 'string'
                }
            },
            'parsingStrategy': 'BEDROCK_FOUNDATION_MODEL'
        }
    }
)

Parameters:

clientToken (string) –
A unique, case-sensitive identifier that you provide to ensure the idempotency of the request. If not provided, the Amazon Web Services SDK populates this field. For more information about idempotency, see Making retries safe with idempotent APIs.

This field is autopopulated if not provided.
description (string) – The description.
knowledgeBaseType (string) –
[REQUIRED]

The type of knowledge base. Only CUSTOM knowledge bases allow you to upload your own content. EXTERNAL knowledge bases support integrations with third-party systems whose content is synchronized automatically.
name (string) –
[REQUIRED]

The name of the knowledge base.
renderingConfiguration (dict) –
Information about how to render the content.
- templateUri (string) –
  
  A URI template containing exactly one variable in ${variableName} ``format. This can only be set for ``EXTERNAL knowledge bases. For Salesforce, ServiceNow, and Zendesk, the variable must be one of the following:
  - Salesforce: Id, ArticleNumber, VersionNumber, Title, PublishStatus, or IsDeleted
  - ServiceNow: number, short_description, sys_mod_count, workflow_state, or active
  - Zendesk: id, title, updated_at, or draft
  The variable is replaced with the actual value for a piece of content when calling GetContent.
serverSideEncryptionConfiguration (dict) –
The configuration information for the customer managed key used for encryption.

This KMS key must have a policy that allows kms:CreateGrant, kms:DescribeKey, kms:Decrypt, and kms:GenerateDataKey* permissions to the IAM identity using the key to invoke Amazon Q in Connect.

For more information about setting up a customer managed key for Amazon Q in Connect, see Enable Amazon Q in Connect for your instance.
- kmsKeyId (string) –
  
  The customer managed key used for encryption. For more information about setting up a customer managed key for Amazon Q in Connect, see Enable Amazon Q in Connect for your instance. For information about valid ID values, see Key identifiers (KeyId).
sourceConfiguration (dict) –
The source of the knowledge base content. Only set this argument for EXTERNAL or Managed knowledge bases.

Note
This is a Tagged Union structure. Only one of the following top level keys can be set: appIntegrations, managedSourceConfiguration.
- appIntegrations (dict) –
  
  Configuration information for Amazon AppIntegrations to automatically ingest content.
  - appIntegrationArn (string) – [REQUIRED]
    
    The Amazon Resource Name (ARN) of the AppIntegrations DataIntegration to use for ingesting content.
    - For Salesforce, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least Id, ArticleNumber, VersionNumber, Title, PublishStatus, and IsDeleted as source fields.
    - For ServiceNow, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least number, short_description, sys_mod_count, workflow_state, and active as source fields.
    - For Zendesk, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least id, title, updated_at, and draft as source fields.
    - For SharePoint, your AppIntegrations DataIntegration must have a FileConfiguration, including only file extensions that are among docx, pdf, html, htm, and txt.
    - For Amazon S3, the ObjectConfiguration and FileConfiguration of your AppIntegrations DataIntegration must be null. The SourceURI of your DataIntegration must use the following format: s3://your_s3_bucket_name.
    Warning
    The bucket policy of the corresponding S3 bucket must allow the Amazon Web Services principal app-integrations.amazonaws.com to perform s3:ListBucket, s3:GetObject, and s3:GetBucketLocation against the bucket.
  - objectFields (list) –
    
    The fields from the source that are made available to your agents in Amazon Q in Connect. Optional if ObjectConfiguration is included in the provided DataIntegration.
    - For Salesforce, you must include at least Id, ArticleNumber, VersionNumber, Title, PublishStatus, and IsDeleted.
    - For ServiceNow, you must include at least number, short_description, sys_mod_count, workflow_state, and active.
    - For Zendesk, you must include at least id, title, updated_at, and draft.
    Make sure to include additional fields. These fields are indexed and used to source recommendations.
    - (string) –
- managedSourceConfiguration (dict) –
  
  Source configuration for managed resources.
  
  Note
  This is a Tagged Union structure. Only one of the following top level keys can be set: webCrawlerConfiguration.
  - webCrawlerConfiguration (dict) –
    
    Configuration data for web crawler data source.
    - crawlerLimits (dict) –
      
      The configuration of crawl limits for the web URLs.
      - rateLimit (integer) –
        
        Rate of web URLs retrieved per minute.
    - exclusionFilters (list) –
      
      A list of one or more exclusion regular expression patterns to exclude certain URLs. If you specify an inclusion and exclusion filter/pattern and both match a URL, the exclusion filter takes precedence and the web content of the URL isn’t crawled.
      - (string) –
    - inclusionFilters (list) –
      
      A list of one or more inclusion regular expression patterns to include certain URLs. If you specify an inclusion and exclusion filter/pattern and both match a URL, the exclusion filter takes precedence and the web content of the URL isn’t crawled.
      - (string) –
    - scope (string) –
      
      The scope of what is crawled for your URLs. You can choose to crawl only web pages that belong to the same host or primary domain. For example, only web pages that contain the seed URL https://docs.aws.amazon.com/bedrock/latest/userguide/ and no other domains. You can choose to include sub domains in addition to the host or primary domain. For example, web pages that contain aws.amazon.com can also include sub domain docs.aws.amazon.com.
    - urlConfiguration (dict) – [REQUIRED]
      
      The configuration of the URL/URLs for the web content that you want to crawl. You should be authorized to crawl the URLs.
      - seedUrls (list) –
        
        List of URLs for crawling.
        
        (dict) –
        
        A URL for crawling.
        
        url (string) –
        
        URL for crawling
tags (dict) –
The tags used to organize, track, or control access for this resource.
- (string) –
  - (string) –
vectorIngestionConfiguration (dict) –
Contains details about how to ingest the documents in a data source.
- chunkingConfiguration (dict) –
  
  Details about how to chunk the documents in the data source. A chunk refers to an excerpt from a data source that is returned when the knowledge base that it belongs to is queried.
  - chunkingStrategy (string) – [REQUIRED]
    
    Knowledge base can split your source data into chunks. A chunk refers to an excerpt from a data source that is returned when the knowledge base that it belongs to is queried. You have the following options for chunking your data. If you opt for NONE, then you may want to pre-process your files by splitting them up such that each file corresponds to a chunk.
  - fixedSizeChunkingConfiguration (dict) –
    
    Configurations for when you choose fixed-size chunking. If you set the chunkingStrategy as NONE, exclude this field.
    - maxTokens (integer) – [REQUIRED]
      
      The maximum number of tokens to include in a chunk.
    - overlapPercentage (integer) – [REQUIRED]
      
      The percentage of overlap between adjacent chunks of a data source.
  - hierarchicalChunkingConfiguration (dict) –
    
    Settings for hierarchical document chunking for a data source. Hierarchical chunking splits documents into layers of chunks where the first layer contains large chunks, and the second layer contains smaller chunks derived from the first layer.
    - levelConfigurations (list) – [REQUIRED]
      
      Token settings for each layer.
      - (dict) –
        
        Token settings for each layer.
        
        maxTokens (integer) – [REQUIRED]
        
        The maximum number of tokens that a chunk can contain in this layer.
    - overlapTokens (integer) – [REQUIRED]
      
      The number of tokens to repeat across chunks in the same layer.
  - semanticChunkingConfiguration (dict) –
    
    Settings for semantic document chunking for a data source. Semantic chunking splits a document into smaller documents based on groups of similar content derived from the text with natural language processing.
    - breakpointPercentileThreshold (integer) – [REQUIRED]
      
      The dissimilarity threshold for splitting chunks.
    - bufferSize (integer) – [REQUIRED]
      
      The buffer size.
    - maxTokens (integer) – [REQUIRED]
      
      The maximum number of tokens that a chunk can contain.
- parsingConfiguration (dict) –
  
  A custom parser for data source documents.
  - bedrockFoundationModelConfiguration (dict) –
    
    Settings for a foundation model used to parse documents for a data source.
    - modelArn (string) – [REQUIRED]
      
      The ARN of the foundation model.
    - parsingPrompt (dict) –
      
      Instructions for interpreting the contents of a document.
      - parsingPromptText (string) – [REQUIRED]
        
        Instructions for interpreting the contents of a document.
  - parsingStrategy (string) – [REQUIRED]
    
    The parsing strategy for the data source.

Return type:

dict

Returns:

Response Syntax

{
    'knowledgeBase': {
        'description': 'string',
        'ingestionFailureReasons': [
            'string',
        ],
        'ingestionStatus': 'SYNC_FAILED'|'SYNCING_IN_PROGRESS'|'SYNC_SUCCESS'|'CREATE_IN_PROGRESS',
        'knowledgeBaseArn': 'string',
        'knowledgeBaseId': 'string',
        'knowledgeBaseType': 'EXTERNAL'|'CUSTOM'|'QUICK_RESPONSES'|'MESSAGE_TEMPLATES'|'MANAGED',
        'lastContentModificationTime': datetime(2015, 1, 1),
        'name': 'string',
        'renderingConfiguration': {
            'templateUri': 'string'
        },
        'serverSideEncryptionConfiguration': {
            'kmsKeyId': 'string'
        },
        'sourceConfiguration': {
            'appIntegrations': {
                'appIntegrationArn': 'string',
                'objectFields': [
                    'string',
                ]
            },
            'managedSourceConfiguration': {
                'webCrawlerConfiguration': {
                    'crawlerLimits': {
                        'rateLimit': 123
                    },
                    'exclusionFilters': [
                        'string',
                    ],
                    'inclusionFilters': [
                        'string',
                    ],
                    'scope': 'HOST_ONLY'|'SUBDOMAINS',
                    'urlConfiguration': {
                        'seedUrls': [
                            {
                                'url': 'string'
                            },
                        ]
                    }
                }
            }
        },
        'status': 'CREATE_IN_PROGRESS'|'CREATE_FAILED'|'ACTIVE'|'DELETE_IN_PROGRESS'|'DELETE_FAILED'|'DELETED',
        'tags': {
            'string': 'string'
        },
        'vectorIngestionConfiguration': {
            'chunkingConfiguration': {
                'chunkingStrategy': 'FIXED_SIZE'|'NONE'|'HIERARCHICAL'|'SEMANTIC',
                'fixedSizeChunkingConfiguration': {
                    'maxTokens': 123,
                    'overlapPercentage': 123
                },
                'hierarchicalChunkingConfiguration': {
                    'levelConfigurations': [
                        {
                            'maxTokens': 123
                        },
                    ],
                    'overlapTokens': 123
                },
                'semanticChunkingConfiguration': {
                    'breakpointPercentileThreshold': 123,
                    'bufferSize': 123,
                    'maxTokens': 123
                }
            },
            'parsingConfiguration': {
                'bedrockFoundationModelConfiguration': {
                    'modelArn': 'string',
                    'parsingPrompt': {
                        'parsingPromptText': 'string'
                    }
                },
                'parsingStrategy': 'BEDROCK_FOUNDATION_MODEL'
            }
        }
    }
}

Response Structure

(dict) –
- knowledgeBase (dict) –
  
  The knowledge base.
  - description (string) –
    
    The description.
  - ingestionFailureReasons (list) –
    
    List of failure reasons on ingestion per file.
    - (string) –
  - ingestionStatus (string) –
    
    Status of ingestion on data source.
  - knowledgeBaseArn (string) –
    
    The Amazon Resource Name (ARN) of the knowledge base.
  - knowledgeBaseId (string) –
    
    The identifier of the knowledge base.
  - knowledgeBaseType (string) –
    
    The type of knowledge base.
  - lastContentModificationTime (datetime) –
    
    An epoch timestamp indicating the most recent content modification inside the knowledge base. If no content exists in a knowledge base, this value is unset.
  - name (string) –
    
    The name of the knowledge base.
  - renderingConfiguration (dict) –
    
    Information about how to render the content.
    - templateUri (string) –
      
      A URI template containing exactly one variable in ${variableName} ``format. This can only be set for ``EXTERNAL knowledge bases. For Salesforce, ServiceNow, and Zendesk, the variable must be one of the following:
      - Salesforce: Id, ArticleNumber, VersionNumber, Title, PublishStatus, or IsDeleted
      - ServiceNow: number, short_description, sys_mod_count, workflow_state, or active
      - Zendesk: id, title, updated_at, or draft
      The variable is replaced with the actual value for a piece of content when calling GetContent.
  - serverSideEncryptionConfiguration (dict) –
    
    The configuration information for the customer managed key used for encryption.
    
    This KMS key must have a policy that allows kms:CreateGrant, kms:DescribeKey, kms:Decrypt, and kms:GenerateDataKey* permissions to the IAM identity using the key to invoke Amazon Q in Connect.
    
    For more information about setting up a customer managed key for Amazon Q in Connect, see Enable Amazon Q in Connect for your instance.
    - kmsKeyId (string) –
      
      The customer managed key used for encryption. For more information about setting up a customer managed key for Amazon Q in Connect, see Enable Amazon Q in Connect for your instance. For information about valid ID values, see Key identifiers (KeyId).
  - sourceConfiguration (dict) –
    
    Source configuration information about the knowledge base.
    Note
    This is a Tagged Union structure. Only one of the following top level keys will be set: appIntegrations, managedSourceConfiguration. If a client receives an unknown member it will set SDK_UNKNOWN_MEMBER as the top level key, which maps to the name or tag of the unknown member. The structure of SDK_UNKNOWN_MEMBER is as follows:
    'SDK_UNKNOWN_MEMBER': {'name': 'UnknownMemberName'}
    - appIntegrations (dict) –
      
      Configuration information for Amazon AppIntegrations to automatically ingest content.
      - appIntegrationArn (string) –
        
        The Amazon Resource Name (ARN) of the AppIntegrations DataIntegration to use for ingesting content.
        
        For Salesforce, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least Id, ArticleNumber, VersionNumber, Title, PublishStatus, and IsDeleted as source fields.
        
        For ServiceNow, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least number, short_description, sys_mod_count, workflow_state, and active as source fields.
        
        For Zendesk, your AppIntegrations DataIntegration must have an ObjectConfiguration if objectFields is not provided, including at least id, title, updated_at, and draft as source fields.
        
        For SharePoint, your AppIntegrations DataIntegration must have a FileConfiguration, including only file extensions that are among docx, pdf, html, htm, and txt.
        
        For Amazon S3, the ObjectConfiguration and FileConfiguration of your AppIntegrations DataIntegration must be null. The SourceURI of your DataIntegration must use the following format: s3://your_s3_bucket_name.
        
        Warning
        The bucket policy of the corresponding S3 bucket must allow the Amazon Web Services principal app-integrations.amazonaws.com to perform s3:ListBucket, s3:GetObject, and s3:GetBucketLocation against the bucket.
      - objectFields (list) –
        
        The fields from the source that are made available to your agents in Amazon Q in Connect. Optional if ObjectConfiguration is included in the provided DataIntegration.
        
        For Salesforce, you must include at least Id, ArticleNumber, VersionNumber, Title, PublishStatus, and IsDeleted.
        
        For ServiceNow, you must include at least number, short_description, sys_mod_count, workflow_state, and active.
        
        For Zendesk, you must include at least id, title, updated_at, and draft.
        
        Make sure to include additional fields. These fields are indexed and used to source recommendations.
        
        (string) –
    - managedSourceConfiguration (dict) –
      
      Source configuration for managed resources.
      Note
      This is a Tagged Union structure. Only one of the following top level keys will be set: webCrawlerConfiguration. If a client receives an unknown member it will set SDK_UNKNOWN_MEMBER as the top level key, which maps to the name or tag of the unknown member. The structure of SDK_UNKNOWN_MEMBER is as follows:
      
      'SDK_UNKNOWN_MEMBER': {'name': 'UnknownMemberName'}
      - webCrawlerConfiguration (dict) –
        
        Configuration data for web crawler data source.
        
        crawlerLimits (dict) –
        
        The configuration of crawl limits for the web URLs.
        
        rateLimit (integer) –
        
        Rate of web URLs retrieved per minute.
        
        exclusionFilters (list) –
        
        A list of one or more exclusion regular expression patterns to exclude certain URLs. If you specify an inclusion and exclusion filter/pattern and both match a URL, the exclusion filter takes precedence and the web content of the URL isn’t crawled.
        
        (string) –
        
        inclusionFilters (list) –
        
        A list of one or more inclusion regular expression patterns to include certain URLs. If you specify an inclusion and exclusion filter/pattern and both match a URL, the exclusion filter takes precedence and the web content of the URL isn’t crawled.
        
        (string) –
        
        scope (string) –
        
        The scope of what is crawled for your URLs. You can choose to crawl only web pages that belong to the same host or primary domain. For example, only web pages that contain the seed URL https://docs.aws.amazon.com/bedrock/latest/userguide/ and no other domains. You can choose to include sub domains in addition to the host or primary domain. For example, web pages that contain aws.amazon.com can also include sub domain docs.aws.amazon.com.
        
        urlConfiguration (dict) –
        
        The configuration of the URL/URLs for the web content that you want to crawl. You should be authorized to crawl the URLs.
        
        seedUrls (list) –
        
        List of URLs for crawling.
        
        (dict) –
        
        A URL for crawling.
        
        url (string) –
        
        URL for crawling
  - status (string) –
    
    The status of the knowledge base.
  - tags (dict) –
    
    The tags used to organize, track, or control access for this resource.
    - (string) –
      - (string) –
  - vectorIngestionConfiguration (dict) –
    
    Contains details about how to ingest the documents in a data source.
    - chunkingConfiguration (dict) –
      
      Details about how to chunk the documents in the data source. A chunk refers to an excerpt from a data source that is returned when the knowledge base that it belongs to is queried.
      - chunkingStrategy (string) –
        
        Knowledge base can split your source data into chunks. A chunk refers to an excerpt from a data source that is returned when the knowledge base that it belongs to is queried. You have the following options for chunking your data. If you opt for NONE, then you may want to pre-process your files by splitting them up such that each file corresponds to a chunk.
      - fixedSizeChunkingConfiguration (dict) –
        
        Configurations for when you choose fixed-size chunking. If you set the chunkingStrategy as NONE, exclude this field.
        
        maxTokens (integer) –
        
        The maximum number of tokens to include in a chunk.
        
        overlapPercentage (integer) –
        
        The percentage of overlap between adjacent chunks of a data source.
      - hierarchicalChunkingConfiguration (dict) –
        
        Settings for hierarchical document chunking for a data source. Hierarchical chunking splits documents into layers of chunks where the first layer contains large chunks, and the second layer contains smaller chunks derived from the first layer.
        
        levelConfigurations (list) –
        
        Token settings for each layer.
        
        (dict) –
        
        Token settings for each layer.
        
        maxTokens (integer) –
        
        The maximum number of tokens that a chunk can contain in this layer.
        
        overlapTokens (integer) –
        
        The number of tokens to repeat across chunks in the same layer.
      - semanticChunkingConfiguration (dict) –
        
        Settings for semantic document chunking for a data source. Semantic chunking splits a document into smaller documents based on groups of similar content derived from the text with natural language processing.
        
        breakpointPercentileThreshold (integer) –
        
        The dissimilarity threshold for splitting chunks.
        
        bufferSize (integer) –
        
        The buffer size.
        
        maxTokens (integer) –
        
        The maximum number of tokens that a chunk can contain.
    - parsingConfiguration (dict) –
      
      A custom parser for data source documents.
      - bedrockFoundationModelConfiguration (dict) –
        
        Settings for a foundation model used to parse documents for a data source.
        
        modelArn (string) –
        
        The ARN of the foundation model.
        
        parsingPrompt (dict) –
        
        Instructions for interpreting the contents of a document.
        
        parsingPromptText (string) –
        
        Instructions for interpreting the contents of a document.
      - parsingStrategy (string) –
        
        The parsing strategy for the data source.

Exceptions

QConnect.Client.exceptions.ConflictException
QConnect.Client.exceptions.ValidationException
QConnect.Client.exceptions.ServiceQuotaExceededException
QConnect.Client.exceptions.AccessDeniedException