Agents for Amazon Bedrock Runtime¶

Description¶

Amazon Bedrock Agents (now Amazon Bedrock Agents Classic) is no longer open to new customers. For capabilities similar to Bedrock Agents Classic, explore Amazon Bedrock AgentCore. Existing customers can continue to use the service as normal. For more information, see Amazon Bedrock Agents Classic availability change.

Contains APIs related to model invocation and querying of knowledge bases.

Usage¶

bedrockagentruntime(
  config = list(),
  credentials = list(),
  endpoint = NULL,
  region = NULL
)

Arguments¶

config

Optional configuration of credentials, endpoint, and/or region.
- credentials:
  - creds:
    - access_key_id: AWS access key ID
    - secret_access_key: AWS secret access key
    - session_token: AWS temporary session token
  - profile: The name of a profile to use. If not given, then the default profile is used.
  - anonymous: Set anonymous credentials.
- endpoint: The complete URL to use for the constructed client.
- region: The AWS Region used in instantiating the client.
- close_connection: Immediately close all HTTP connections.
- timeout: The time in seconds till a timeout exception is thrown when attempting to make a connection. The default is 60 seconds.
- s3_force_path_style: Set this to true to force the request to use path-style addressing, i.e. http://s3.amazonaws.com/BUCKET/KEY.
- sts_regional_endpoint: Set sts regional endpoint resolver to regional or legacy https://docs.aws.amazon.com/sdkref/latest/guide/feature-sts-regionalized-endpoints.html
- use_dual_stack: Set this to true to use the dualstack (IPv4 and IPv6) endpoint for a service, where available, falling back to the regular endpoint when it isn't. Defaults to the AWS_USE_DUALSTACK_ENDPOINT environment variable when unset.
credentials

Optional credentials shorthand for the config parameter
- creds:
  - access_key_id: AWS access key ID
  - secret_access_key: AWS secret access key
  - session_token: AWS temporary session token
- profile: The name of a profile to use. If not given, then the default profile is used.
- anonymous: Set anonymous credentials.
endpoint

Optional shorthand for complete URL to use for the constructed client.
region

Optional shorthand for AWS Region used in instantiating the client.

Value¶

A client for the service. You can call the service's operations using syntax like svc$operation(...), where svc is the name you've assigned to the client. The available operations are listed in the Operations section.

Service syntax¶

svc <- bedrockagentruntime(
  config = list(
    credentials = list(
      creds = list(
        access_key_id = "string",
        secret_access_key = "string",
        session_token = "string"
      ),
      profile = "string",
      anonymous = "logical"
    ),
    endpoint = "string",
    region = "string",
    close_connection = "logical",
    timeout = "numeric",
    s3_force_path_style = "logical",
    sts_regional_endpoint = "string",
    use_dual_stack = "logical"
  ),
  credentials = list(
    creds = list(
      access_key_id = "string",
      secret_access_key = "string",
      session_token = "string"
    ),
    profile = "string",
    anonymous = "logical"
  ),
  endpoint = "string",
  region = "string"
)

Operations¶

`agentic_retrieve_stream`	Retrieves information from one or more knowledge bases using an agentic approach
`check_ingested_document_acl`	Checks whether a user has access to a specific document by verifying against the ingested access control list (ACL) in a knowledge base
`create_invocation`	Creates a new invocation within a session
`create_session`	Creates a session to temporarily store conversations for generative AI (GenAI) applications built with open-source frameworks such as LangGraph and LlamaIndex
`delete_agent_memory`	Deletes memory from the specified memory identifier
`delete_session`	Deletes a session that you ended
`end_session`	Ends the session
`generate_query`	Generates an SQL query from a natural language query
`get_agent_memory`	Gets the sessions stored in the memory of the agent
`get_document_content`	Retrieves the content of an ingested document from a knowledge base
`get_execution_flow_snapshot`	Retrieves the flow definition snapshot used for a flow execution
`get_flow_execution`	Retrieves details about a specific flow execution, including its status, start and end times, and any errors that occurred during execution
`get_ingested_document_acl`	Retrieves the ingested access control list (ACL) for a specific document in a knowledge base
`get_invocation_step`	Retrieves the details of a specific invocation step within an invocation in a session
`get_session`	Retrieves details about a specific session
`invoke_agent`	Amazon Bedrock Agents (now Amazon Bedrock Agents Classic) is no longer open to new customers
`invoke_flow`	Invokes an alias of a flow to run the inputs that you specify and return the output of each node as a stream
`invoke_inline_agent`	Invokes an inline Amazon Bedrock agent using the configurations you provide with the request
`list_flow_execution_events`	Lists events that occurred during a flow execution
`list_flow_executions`	Lists all executions of a flow
`list_invocations`	Lists all invocations associated with a specific session
`list_invocation_steps`	Lists all invocation steps associated with a session and optionally, an invocation within the session
`list_sessions`	Lists all sessions in your Amazon Web Services account
`list_tags_for_resource`	List all the tags for the resource you specify
`optimize_prompt`	Optimizes a prompt for the task that you specify
`put_invocation_step`	Add an invocation step to an invocation in a session
`rerank`	Reranks the relevance of sources based on queries
`retrieve`	Queries a knowledge base and retrieves information from it
`retrieve_and_generate`	Queries a knowledge base and generates responses based on the retrieved results and using the specified foundation model or inference profile
`retrieve_and_generate_stream`	Queries a knowledge base and generates responses based on the retrieved results, with output in streaming format
`start_flow_execution`	Starts an execution of an Amazon Bedrock flow
`stop_flow_execution`	Stops an Amazon Bedrock flow's execution
`tag_resource`	Associate tags with a resource
`untag_resource`	Remove tags from a resource
`update_session`	Updates the metadata or encryption settings of a session

Examples¶

svc <- bedrockagentruntime()
svc$agentic_retrieve_stream(
  Foo = 123
)