Class: Aws::BedrockAgentRuntime::Client
- Inherits:
-
Seahorse::Client::Base
- Object
- Seahorse::Client::Base
- Aws::BedrockAgentRuntime::Client
- Includes:
- ClientStubs
- Defined in:
- gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb
Overview
An API client for BedrockAgentRuntime. To construct a client, you need to configure a :region
and :credentials
.
client = Aws::BedrockAgentRuntime::Client.new(
region: region_name,
credentials: credentials,
# ...
)
For details on configuring region and credentials see the developer guide.
See #initialize for a full list of supported configuration options.
Instance Attribute Summary
Attributes inherited from Seahorse::Client::Base
API Operations collapse
-
#delete_agent_memory(params = {}) ⇒ Struct
Deletes memory from the specified memory identifier.
-
#get_agent_memory(params = {}) ⇒ Types::GetAgentMemoryResponse
Gets the sessions stored in the memory of the agent.
-
#invoke_agent(params = {}) ⇒ Types::InvokeAgentResponse
The CLI doesn't support streaming operations in Amazon Bedrock, including InvokeAgent
. -
#invoke_flow(params = {}) ⇒ Types::InvokeFlowResponse
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(params = {}) ⇒ Types::InvokeInlineAgentResponse
Invokes an inline Amazon Bedrock agent using the configurations you provide with the request.
-
#optimize_prompt(params = {}) ⇒ Types::OptimizePromptResponse
Optimizes a prompt for the task that you specify.
-
#retrieve(params = {}) ⇒ Types::RetrieveResponse
Queries a knowledge base and retrieves information from it.
-
#retrieve_and_generate(params = {}) ⇒ Types::RetrieveAndGenerateResponse
Queries a knowledge base and generates responses based on the retrieved results and using the specified foundation model or [inference profile][1].
Instance Method Summary collapse
-
#initialize(options) ⇒ Client
constructor
A new instance of Client.
Methods included from ClientStubs
#api_requests, #stub_data, #stub_responses
Methods inherited from Seahorse::Client::Base
add_plugin, api, clear_plugins, define, new, #operation_names, plugins, remove_plugin, set_api, set_plugins
Methods included from Seahorse::Client::HandlerBuilder
#handle, #handle_request, #handle_response
Constructor Details
#initialize(options) ⇒ Client
Returns a new instance of Client.
Parameters:
- options (Hash)
Options Hash (options):
-
:plugins
(Array<Seahorse::Client::Plugin>)
— default:
[]]
—
A list of plugins to apply to the client. Each plugin is either a class name or an instance of a plugin class.
-
:credentials
(required, Aws::CredentialProvider)
—
Your AWS credentials. This can be an instance of any one of the following classes:
Aws::Credentials
- Used for configuring static, non-refreshing credentials.Aws::SharedCredentials
- Used for loading static credentials from a shared file, such as~/.aws/config
.Aws::AssumeRoleCredentials
- Used when you need to assume a role.Aws::AssumeRoleWebIdentityCredentials
- Used when you need to assume a role after providing credentials via the web.Aws::SSOCredentials
- Used for loading credentials from AWS SSO using an access token generated fromaws login
.Aws::ProcessCredentials
- Used for loading credentials from a process that outputs to stdout.Aws::InstanceProfileCredentials
- Used for loading credentials from an EC2 IMDS on an EC2 instance.Aws::ECSCredentials
- Used for loading credentials from instances running in ECS.Aws::CognitoIdentityCredentials
- Used for loading credentials from the Cognito Identity service.
When
:credentials
are not configured directly, the following locations will be searched for credentials:Aws.config[:credentials]
- The
:access_key_id
,:secret_access_key
,:session_token
, and:account_id
options. - ENV['AWS_ACCESS_KEY_ID'], ENV['AWS_SECRET_ACCESS_KEY'], ENV['AWS_SESSION_TOKEN'], and ENV['AWS_ACCOUNT_ID']
~/.aws/credentials
~/.aws/config
- EC2/ECS IMDS instance profile - When used by default, the timeouts
are very aggressive. Construct and pass an instance of
Aws::InstanceProfileCredentials
orAws::ECSCredentials
to enable retries and extended timeouts. Instance profile credential fetching can be disabled by setting ENV['AWS_EC2_METADATA_DISABLED'] to true.
-
:region
(required, String)
—
The AWS region to connect to. The configured
:region
is used to determine the service:endpoint
. When not passed, a default:region
is searched for in the following locations:Aws.config[:region]
ENV['AWS_REGION']
ENV['AMAZON_REGION']
ENV['AWS_DEFAULT_REGION']
~/.aws/credentials
~/.aws/config
- :access_key_id (String)
- :account_id (String)
-
:active_endpoint_cache
(Boolean)
— default:
false
—
When set to
true
, a thread polling for endpoints will be running in the background every 60 secs (default). Defaults tofalse
. -
:adaptive_retry_wait_to_fill
(Boolean)
— default:
true
—
Used only in
adaptive
retry mode. When true, the request will sleep until there is sufficent client side capacity to retry the request. When false, the request will raise aRetryCapacityNotAvailableError
and will not retry instead of sleeping. -
:client_side_monitoring
(Boolean)
— default:
false
—
When
true
, client-side metrics will be collected for all API requests from this client. -
:client_side_monitoring_client_id
(String)
— default:
""
—
Allows you to provide an identifier for this client which will be attached to all generated client side metrics. Defaults to an empty string.
-
:client_side_monitoring_host
(String)
— default:
"127.0.0.1"
—
Allows you to specify the DNS hostname or IPv4 or IPv6 address that the client side monitoring agent is running on, where client metrics will be published via UDP.
-
:client_side_monitoring_port
(Integer)
— default:
31000
—
Required for publishing client metrics. The port that the client side monitoring agent is running on, where client metrics will be published via UDP.
-
:client_side_monitoring_publisher
(Aws::ClientSideMonitoring::Publisher)
— default:
Aws::ClientSideMonitoring::Publisher
—
Allows you to provide a custom client-side monitoring publisher class. By default, will use the Client Side Monitoring Agent Publisher.
-
:convert_params
(Boolean)
— default:
true
—
When
true
, an attempt is made to coerce request parameters into the required types. -
:correct_clock_skew
(Boolean)
— default:
true
—
Used only in
standard
and adaptive retry modes. Specifies whether to apply a clock skew correction and retry requests with skewed client clocks. -
:defaults_mode
(String)
— default:
"legacy"
—
See DefaultsModeConfiguration for a list of the accepted modes and the configuration defaults that are included.
-
:disable_host_prefix_injection
(Boolean)
— default:
false
—
Set to true to disable SDK automatically adding host prefix to default service endpoint when available.
-
:disable_request_compression
(Boolean)
— default:
false
—
When set to 'true' the request body will not be compressed for supported operations.
-
:endpoint
(String, URI::HTTPS, URI::HTTP)
—
Normally you should not configure the
:endpoint
option directly. This is normally constructed from the:region
option. Configuring:endpoint
is normally reserved for connecting to test or custom endpoints. The endpoint should be a URI formatted like:'http://example.com' 'https://example.com' 'http://example.com:123'
-
:endpoint_cache_max_entries
(Integer)
— default:
1000
—
Used for the maximum size limit of the LRU cache storing endpoints data for endpoint discovery enabled operations. Defaults to 1000.
-
:endpoint_cache_max_threads
(Integer)
— default:
10
—
Used for the maximum threads in use for polling endpoints to be cached, defaults to 10.
-
:endpoint_cache_poll_interval
(Integer)
— default:
60
—
When :endpoint_discovery and :active_endpoint_cache is enabled, Use this option to config the time interval in seconds for making requests fetching endpoints information. Defaults to 60 sec.
-
:endpoint_discovery
(Boolean)
— default:
false
—
When set to
true
, endpoint discovery will be enabled for operations when available. -
:event_stream_handler
(Proc)
—
When an EventStream or Proc object is provided, it will be used as callback for each chunk of event stream response received along the way.
-
:ignore_configured_endpoint_urls
(Boolean)
—
Setting to true disables use of endpoint URLs provided via environment variables and the shared configuration file.
-
:input_event_stream_handler
(Proc)
—
When an EventStream or Proc object is provided, it can be used for sending events for the event stream.
-
:log_formatter
(Aws::Log::Formatter)
— default:
Aws::Log::Formatter.default
—
The log formatter.
-
:log_level
(Symbol)
— default:
:info
—
The log level to send messages to the
:logger
at. -
:logger
(Logger)
—
The Logger instance to send log messages to. If this option is not set, logging will be disabled.
-
:max_attempts
(Integer)
— default:
3
—
An integer representing the maximum number attempts that will be made for a single request, including the initial attempt. For example, setting this value to 5 will result in a request being retried up to 4 times. Used in
standard
andadaptive
retry modes. -
:output_event_stream_handler
(Proc)
—
When an EventStream or Proc object is provided, it will be used as callback for each chunk of event stream response received along the way.
-
:profile
(String)
— default:
"default"
—
Used when loading credentials from the shared credentials file at HOME/.aws/credentials. When not specified, 'default' is used.
-
:request_min_compression_size_bytes
(Integer)
— default:
10240
—
The minimum size in bytes that triggers compression for request bodies. The value must be non-negative integer value between 0 and 10485780 bytes inclusive.
-
:retry_backoff
(Proc)
—
A proc or lambda used for backoff. Defaults to 2**retries * retry_base_delay. This option is only used in the
legacy
retry mode. -
:retry_base_delay
(Float)
— default:
0.3
—
The base delay in seconds used by the default backoff function. This option is only used in the
legacy
retry mode. -
:retry_jitter
(Symbol)
— default:
:none
—
A delay randomiser function used by the default backoff function. Some predefined functions can be referenced by name - :none, :equal, :full, otherwise a Proc that takes and returns a number. This option is only used in the
legacy
retry mode.@see https://www.awsarchitectureblog.com/2015/03/backoff.html
-
:retry_limit
(Integer)
— default:
3
—
The maximum number of times to retry failed requests. Only ~ 500 level server errors and certain ~ 400 level client errors are retried. Generally, these are throttling errors, data checksum errors, networking errors, timeout errors, auth errors, endpoint discovery, and errors from expired credentials. This option is only used in the
legacy
retry mode. -
:retry_max_delay
(Integer)
— default:
0
—
The maximum number of seconds to delay between retries (0 for no limit) used by the default backoff function. This option is only used in the
legacy
retry mode. -
:retry_mode
(String)
— default:
"legacy"
—
Specifies which retry algorithm to use. Values are:
legacy
- The pre-existing retry behavior. This is default value if no retry mode is provided.standard
- A standardized set of retry rules across the AWS SDKs. This includes support for retry quotas, which limit the number of unsuccessful retries a client can make.adaptive
- An experimental retry mode that includes all the functionality ofstandard
mode along with automatic client side throttling. This is a provisional mode that may change behavior in the future.
-
:sdk_ua_app_id
(String)
—
A unique and opaque application ID that is appended to the User-Agent header as app/sdk_ua_app_id. It should have a maximum length of 50. This variable is sourced from environment variable AWS_SDK_UA_APP_ID or the shared config profile attribute sdk_ua_app_id.
- :secret_access_key (String)
- :session_token (String)
-
:sigv4a_signing_region_set
(Array)
—
A list of regions that should be signed with SigV4a signing. When not passed, a default
:sigv4a_signing_region_set
is searched for in the following locations:Aws.config[:sigv4a_signing_region_set]
ENV['AWS_SIGV4A_SIGNING_REGION_SET']
~/.aws/config
-
:stub_responses
(Boolean)
— default:
false
—
Causes the client to return stubbed responses. By default fake responses are generated and returned. You can specify the response data to return or errors to raise by calling ClientStubs#stub_responses. See ClientStubs for more information.
Please note When response stubbing is enabled, no HTTP requests are made, and retries are disabled.
-
:telemetry_provider
(Aws::Telemetry::TelemetryProviderBase)
— default:
Aws::Telemetry::NoOpTelemetryProvider
—
Allows you to provide a telemetry provider, which is used to emit telemetry data. By default, uses
NoOpTelemetryProvider
which will not record or emit any telemetry data. The SDK supports the following telemetry providers:- OpenTelemetry (OTel) - To use the OTel provider, install and require the
opentelemetry-sdk
gem and then, pass in an instance of aAws::Telemetry::OTelProvider
for telemetry provider.
- OpenTelemetry (OTel) - To use the OTel provider, install and require the
-
:token_provider
(Aws::TokenProvider)
—
A Bearer Token Provider. This can be an instance of any one of the following classes:
Aws::StaticTokenProvider
- Used for configuring static, non-refreshing tokens.Aws::SSOTokenProvider
- Used for loading tokens from AWS SSO using an access token generated fromaws login
.
When
:token_provider
is not configured directly, theAws::TokenProviderChain
will be used to search for tokens configured for your profile in shared configuration files. -
:use_dualstack_endpoint
(Boolean)
—
When set to
true
, dualstack enabled endpoints (with.aws
TLD) will be used if available. -
:use_fips_endpoint
(Boolean)
—
When set to
true
, fips compatible endpoints will be used if available. When afips
region is used, the region is normalized and this config is set totrue
. -
:validate_params
(Boolean)
— default:
true
—
When
true
, request parameters are validated before sending the request. -
:endpoint_provider
(Aws::BedrockAgentRuntime::EndpointProvider)
—
The endpoint provider used to resolve endpoints. Any object that responds to
#resolve_endpoint(parameters)
whereparameters
is a Struct similar toAws::BedrockAgentRuntime::EndpointParameters
. -
:http_continue_timeout
(Float)
— default:
1
—
The number of seconds to wait for a 100-continue response before sending the request body. This option has no effect unless the request has "Expect" header set to "100-continue". Defaults to
nil
which disables this behaviour. This value can safely be set per request on the session. -
:http_idle_timeout
(Float)
— default:
5
—
The number of seconds a connection is allowed to sit idle before it is considered stale. Stale connections are closed and removed from the pool before making a request.
-
:http_open_timeout
(Float)
— default:
15
—
The default number of seconds to wait for response data. This value can safely be set per-request on the session.
-
:http_proxy
(URI::HTTP, String)
—
A proxy to send requests through. Formatted like 'http://proxy.com:123'.
-
:http_read_timeout
(Float)
— default:
60
—
The default number of seconds to wait for response data. This value can safely be set per-request on the session.
-
:http_wire_trace
(Boolean)
— default:
false
—
When
true
, HTTP debug output will be sent to the:logger
. -
:on_chunk_received
(Proc)
—
When a Proc object is provided, it will be used as callback when each chunk of the response body is received. It provides three arguments: the chunk, the number of bytes received, and the total number of bytes in the response (or nil if the server did not send a
content-length
). -
:on_chunk_sent
(Proc)
—
When a Proc object is provided, it will be used as callback when each chunk of the request body is sent. It provides three arguments: the chunk, the number of bytes read from the body, and the total number of bytes in the body.
-
:raise_response_errors
(Boolean)
— default:
true
—
When
true
, response errors are raised. -
:ssl_ca_bundle
(String)
—
Full path to the SSL certificate authority bundle file that should be used when verifying peer certificates. If you do not pass
:ssl_ca_bundle
or:ssl_ca_directory
the the system default will be used if available. -
:ssl_ca_directory
(String)
—
Full path of the directory that contains the unbundled SSL certificate authority files for verifying peer certificates. If you do not pass
:ssl_ca_bundle
or:ssl_ca_directory
the the system default will be used if available. -
:ssl_ca_store
(String)
—
Sets the X509::Store to verify peer certificate.
-
:ssl_cert
(OpenSSL::X509::Certificate)
—
Sets a client certificate when creating http connections.
-
:ssl_key
(OpenSSL::PKey)
—
Sets a client key when creating http connections.
-
:ssl_timeout
(Float)
—
Sets the SSL timeout in seconds
-
:ssl_verify_peer
(Boolean)
— default:
true
—
When
true
, SSL peer certificates are verified when establishing a connection.
455 456 457 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 455 def initialize(*args) super end |
Instance Method Details
#delete_agent_memory(params = {}) ⇒ Struct
Deletes memory from the specified memory identifier.
Examples:
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.delete_agent_memory({
agent_alias_id: "AgentAliasId", # required
agent_id: "AgentId", # required
memory_id: "MemoryId",
})
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:agent_alias_id
(required, String)
—
The unique identifier of an alias of an agent.
-
:agent_id
(required, String)
—
The unique identifier of the agent to which the alias belongs.
-
:memory_id
(String)
—
The unique identifier of the memory.
Returns:
-
(Struct)
—
Returns an empty response.
See Also:
486 487 488 489 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 486 def delete_agent_memory(params = {}, options = {}) req = build_request(:delete_agent_memory, params) req.send_request(options) end |
#get_agent_memory(params = {}) ⇒ Types::GetAgentMemoryResponse
Gets the sessions stored in the memory of the agent.
The returned response is a pageable response and is Enumerable. For details on usage see PageableResponse.
Examples:
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.get_agent_memory({
agent_alias_id: "AgentAliasId", # required
agent_id: "AgentId", # required
max_items: 1,
memory_id: "MemoryId", # required
memory_type: "SESSION_SUMMARY", # required, accepts SESSION_SUMMARY
next_token: "NextToken",
})
Response structure
Response structure
resp.memory_contents #=> Array
resp.memory_contents[0].session_summary.memory_id #=> String
resp.memory_contents[0].session_summary.session_expiry_time #=> Time
resp.memory_contents[0].session_summary.session_id #=> String
resp.memory_contents[0].session_summary.session_start_time #=> Time
resp.memory_contents[0].session_summary.summary_text #=> String
resp.next_token #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:agent_alias_id
(required, String)
—
The unique identifier of an alias of an agent.
-
:agent_id
(required, String)
—
The unique identifier of the agent to which the alias belongs.
-
:max_items
(Integer)
—
The maximum number of items to return in the response. If the total number of results is greater than this value, use the token returned in the response in the
nextToken
field when making another request to return the next batch of results. -
:memory_id
(required, String)
—
The unique identifier of the memory.
-
:memory_type
(required, String)
—
The type of memory.
-
:next_token
(String)
—
If the total number of results is greater than the maxItems value provided in the request, enter the token returned in the
nextToken
field in the response in this field to return the next batch of results.
Returns:
-
(Types::GetAgentMemoryResponse)
—
Returns a response object which responds to the following methods:
- #memory_contents => Array<Types::Memory>
- #next_token => String
See Also:
549 550 551 552 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 549 def get_agent_memory(params = {}, options = {}) req = build_request(:get_agent_memory, params) req.send_request(options) end |
#invoke_agent(params = {}) ⇒ Types::InvokeAgentResponse
InvokeAgent
.
Sends a prompt for the agent to process and respond to. Note the following fields for the request:
To continue the same conversation with an agent, use the same
sessionId
value in the request.To activate trace enablement, turn
enableTrace
totrue
. Trace enablement helps you follow the agent's reasoning process that led it to the information it processed, the actions it took, and the final result it yielded. For more information, see Trace enablement.End a conversation by setting
endSession
totrue
.In the
sessionState
object, you can include attributes for the session or prompt or, if you configured an action group to return control, results from invocation of the action group.
The response is returned in the bytes
field of the chunk
object.
The
attribution
object contains citations for parts of the response.If you set
enableTrace
totrue
in the request, you can trace the agent's steps and reasoning process that led it to the response.If the action predicted was configured to return control, the response returns parameters for the action, elicited from the user, in the
returnControl
field.Errors are also surfaced in the response.
Examples:
EventStream Operation Example
EventStream Operation Example
You can process the event once it arrives immediately, or wait until the
full response is complete and iterate through the eventstream enumerator.
To interact with event immediately, you need to register #invoke_agent
with callbacks. Callbacks can be registered for specific events or for all
events, including error events.
Callbacks can be passed into the `:event_stream_handler` option or within a
block statement attached to the #invoke_agent call directly. Hybrid
pattern of both is also supported.
`:event_stream_handler` option takes in either a Proc object or
Aws::BedrockAgentRuntime::EventStreams::ResponseStream object.
Usage pattern a): Callbacks with a block attached to #invoke_agent
Example for registering callbacks for all event types and an error event
client.invoke_agent( # params input# ) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
stream.on_event do |event|
# process all events arrive
puts event.event_type
...
end
end
Usage pattern b): Pass in `:event_stream_handler` for #invoke_agent
1) Create a Aws::BedrockAgentRuntime::EventStreams::ResponseStream object
Example for registering callbacks with specific events
handler = Aws::BedrockAgentRuntime::EventStreams::ResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_agent( # params input #, event_stream_handler: handler)
2) Use a Ruby Proc object
Example for registering callbacks with specific events
handler = Proc.new do |stream|
stream.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
stream.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
stream.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
stream.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
stream.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
stream.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
stream.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
stream.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
stream.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
stream.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
stream.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
stream.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
stream.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
end
client.invoke_agent( # params input #, event_stream_handler: handler)
Usage pattern c): Hybrid pattern of a) and b)
handler = Aws::BedrockAgentRuntime::EventStreams::ResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_agent( # params input #, event_stream_handler: handler) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
end
You can also iterate through events after the response complete.
Events are available at resp.completion # => Enumerator
For parameter input example, please refer to following request syntax
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.invoke_agent({
agent_alias_id: "AgentAliasId", # required
agent_id: "AgentId", # required
enable_trace: false,
end_session: false,
input_text: "InputText",
memory_id: "MemoryId",
session_id: "SessionId", # required
session_state: {
files: [
{
name: "String", # required
source: { # required
byte_content: {
data: "data", # required
media_type: "MimeType", # required
},
s3_location: {
uri: "S3Uri", # required
},
source_type: "S3", # required, accepts S3, BYTE_CONTENT
},
use_case: "CODE_INTERPRETER", # required, accepts CODE_INTERPRETER, CHAT
},
],
invocation_id: "String",
knowledge_base_configurations: [
{
knowledge_base_id: "KnowledgeBaseId", # required
retrieval_configuration: { # required
vector_search_configuration: { # required
filter: {
and_all: [
{
# recursive RetrievalFilter
},
],
equals: {
key: "FilterKey", # required
value: { # required
},
},
greater_than: {
key: "FilterKey", # required
value: { # required
},
},
greater_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
in: {
key: "FilterKey", # required
value: { # required
},
},
less_than: {
key: "FilterKey", # required
value: { # required
},
},
less_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
list_contains: {
key: "FilterKey", # required
value: { # required
},
},
not_equals: {
key: "FilterKey", # required
value: { # required
},
},
not_in: {
key: "FilterKey", # required
value: { # required
},
},
or_all: [
{
# recursive RetrievalFilter
},
],
starts_with: {
key: "FilterKey", # required
value: { # required
},
},
string_contains: {
key: "FilterKey", # required
value: { # required
},
},
},
number_of_results: 1,
override_search_type: "HYBRID", # accepts HYBRID, SEMANTIC
},
},
},
],
prompt_session_attributes: {
"String" => "String",
},
return_control_invocation_results: [
{
api_result: {
action_group: "String", # required
api_path: "ApiPath",
confirmation_state: "CONFIRM", # accepts CONFIRM, DENY
http_method: "String",
http_status_code: 1,
response_body: {
"String" => {
body: "String",
},
},
response_state: "FAILURE", # accepts FAILURE, REPROMPT
},
function_result: {
action_group: "String", # required
confirmation_state: "CONFIRM", # accepts CONFIRM, DENY
function: "String",
response_body: {
"String" => {
body: "String",
},
},
response_state: "FAILURE", # accepts FAILURE, REPROMPT
},
},
],
session_attributes: {
"String" => "String",
},
},
})
Response structure
Response structure
All events are available at resp.completion:
resp.completion #=> Enumerator
resp.completion.event_types #=> [:access_denied_exception, :bad_gateway_exception, :chunk, :conflict_exception, :dependency_failed_exception, :files, :internal_server_exception, :resource_not_found_exception, :return_control, :service_quota_exceeded_exception, :throttling_exception, :trace, :validation_exception]
For :access_denied_exception event available at #on_access_denied_exception_event callback and response eventstream enumerator:
event.message #=> String
For :bad_gateway_exception event available at #on_bad_gateway_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :chunk event available at #on_chunk_event callback and response eventstream enumerator:
event.attribution.citations #=> Array
event.attribution.citations[0].generated_response_part.text_response_part.span.end #=> Integer
event.attribution.citations[0].generated_response_part.text_response_part.span.start #=> Integer
event.attribution.citations[0].generated_response_part.text_response_part.text #=> String
event.attribution.citations[0].retrieved_references #=> Array
event.attribution.citations[0].retrieved_references[0].content.text #=> String
event.attribution.citations[0].retrieved_references[0].location.confluence_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.s3_location.uri #=> String
event.attribution.citations[0].retrieved_references[0].location.salesforce_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.share_point_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
event.attribution.citations[0].retrieved_references[0].location.web_location.url #=> String
event.attribution.citations[0].retrieved_references[0].metadata #=> Hash
event.bytes #=> String
For :conflict_exception event available at #on_conflict_exception_event callback and response eventstream enumerator:
event.message #=> String
For :dependency_failed_exception event available at #on_dependency_failed_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :files event available at #on_files_event callback and response eventstream enumerator:
event.files #=> Array
event.files[0].bytes #=> String
event.files[0].name #=> String
event.files[0].type #=> String
For :internal_server_exception event available at #on_internal_server_exception_event callback and response eventstream enumerator:
event.message #=> String
For :resource_not_found_exception event available at #on_resource_not_found_exception_event callback and response eventstream enumerator:
event.message #=> String
For :return_control event available at #on_return_control_event callback and response eventstream enumerator:
event.invocation_id #=> String
event.invocation_inputs #=> Array
event.invocation_inputs[0].api_invocation_input.action_group #=> String
event.invocation_inputs[0].api_invocation_input.action_invocation_type #=> String, one of "RESULT", "USER_CONFIRMATION", "USER_CONFIRMATION_AND_RESULT"
event.invocation_inputs[0].api_invocation_input.api_path #=> String
event.invocation_inputs[0].api_invocation_input.http_method #=> String
event.invocation_inputs[0].api_invocation_input.parameters #=> Array
event.invocation_inputs[0].api_invocation_input.parameters[0].name #=> String
event.invocation_inputs[0].api_invocation_input.parameters[0].type #=> String
event.invocation_inputs[0].api_invocation_input.parameters[0].value #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content #=> Hash
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties #=> Array
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].name #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].type #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].value #=> String
event.invocation_inputs[0].function_invocation_input.action_group #=> String
event.invocation_inputs[0].function_invocation_input.action_invocation_type #=> String, one of "RESULT", "USER_CONFIRMATION", "USER_CONFIRMATION_AND_RESULT"
event.invocation_inputs[0].function_invocation_input.function #=> String
event.invocation_inputs[0].function_invocation_input.parameters #=> Array
event.invocation_inputs[0].function_invocation_input.parameters[0].name #=> String
event.invocation_inputs[0].function_invocation_input.parameters[0].type #=> String
event.invocation_inputs[0].function_invocation_input.parameters[0].value #=> String
For :service_quota_exceeded_exception event available at #on_service_quota_exceeded_exception_event callback and response eventstream enumerator:
event.message #=> String
For :throttling_exception event available at #on_throttling_exception_event callback and response eventstream enumerator:
event.message #=> String
For :trace event available at #on_trace_event callback and response eventstream enumerator:
event.agent_alias_id #=> String
event.agent_id #=> String
event.agent_version #=> String
event.session_id #=> String
event.trace.failure_trace.failure_reason #=> String
event.trace.failure_trace.trace_id #=> String
event.trace.guardrail_trace.action #=> String, one of "INTERVENED", "NONE"
event.trace.guardrail_trace.input_assessments #=> Array
event.trace.guardrail_trace.input_assessments[0].content_policy.filters #=> Array
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].confidence #=> String, one of "NONE", "LOW", "MEDIUM", "HIGH"
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].type #=> String, one of "INSULTS", "HATE", "SEXUAL", "VIOLENCE", "MISCONDUCT", "PROMPT_ATTACK"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities #=> Array
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].type #=> String, one of "ADDRESS", "AGE", "AWS_ACCESS_KEY", "AWS_SECRET_KEY", "CA_HEALTH_NUMBER", "CA_SOCIAL_INSURANCE_NUMBER", "CREDIT_DEBIT_CARD_CVV", "CREDIT_DEBIT_CARD_EXPIRY", "CREDIT_DEBIT_CARD_NUMBER", "DRIVER_ID", "EMAIL", "INTERNATIONAL_BANK_ACCOUNT_NUMBER", "IP_ADDRESS", "LICENSE_PLATE", "MAC_ADDRESS", "NAME", "PASSWORD", "PHONE", "PIN", "SWIFT_CODE", "UK_NATIONAL_HEALTH_SERVICE_NUMBER", "UK_NATIONAL_INSURANCE_NUMBER", "UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER", "URL", "USERNAME", "US_BANK_ACCOUNT_NUMBER", "US_BANK_ROUTING_NUMBER", "US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER", "US_PASSPORT_NUMBER", "US_SOCIAL_SECURITY_NUMBER", "VEHICLE_IDENTIFICATION_NUMBER"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes #=> Array
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].name #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].regex #=> String
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics #=> Array
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].name #=> String
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].type #=> String, one of "DENY"
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words #=> Array
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists #=> Array
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].type #=> String, one of "PROFANITY"
event.trace.guardrail_trace.output_assessments #=> Array
event.trace.guardrail_trace.output_assessments[0].content_policy.filters #=> Array
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].confidence #=> String, one of "NONE", "LOW", "MEDIUM", "HIGH"
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].type #=> String, one of "INSULTS", "HATE", "SEXUAL", "VIOLENCE", "MISCONDUCT", "PROMPT_ATTACK"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities #=> Array
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].type #=> String, one of "ADDRESS", "AGE", "AWS_ACCESS_KEY", "AWS_SECRET_KEY", "CA_HEALTH_NUMBER", "CA_SOCIAL_INSURANCE_NUMBER", "CREDIT_DEBIT_CARD_CVV", "CREDIT_DEBIT_CARD_EXPIRY", "CREDIT_DEBIT_CARD_NUMBER", "DRIVER_ID", "EMAIL", "INTERNATIONAL_BANK_ACCOUNT_NUMBER", "IP_ADDRESS", "LICENSE_PLATE", "MAC_ADDRESS", "NAME", "PASSWORD", "PHONE", "PIN", "SWIFT_CODE", "UK_NATIONAL_HEALTH_SERVICE_NUMBER", "UK_NATIONAL_INSURANCE_NUMBER", "UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER", "URL", "USERNAME", "US_BANK_ACCOUNT_NUMBER", "US_BANK_ROUTING_NUMBER", "US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER", "US_PASSPORT_NUMBER", "US_SOCIAL_SECURITY_NUMBER", "VEHICLE_IDENTIFICATION_NUMBER"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes #=> Array
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].name #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].regex #=> String
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics #=> Array
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].name #=> String
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].type #=> String, one of "DENY"
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words #=> Array
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists #=> Array
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].type #=> String, one of "PROFANITY"
event.trace.guardrail_trace.trace_id #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.action_group_name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.api_path #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.execution_type #=> String, one of "LAMBDA", "RETURN_CONTROL"
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.function #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.invocation_id #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters #=> Array
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].type #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].value #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content #=> Hash
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"] #=> Array
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].type #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].value #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.verb #=> String
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.code #=> String
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.files #=> Array
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.files[0] #=> String
event.trace.orchestration_trace.invocation_input.invocation_type #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "FINISH", "ACTION_GROUP_CODE_INTERPRETER"
event.trace.orchestration_trace.invocation_input.knowledge_base_lookup_input.knowledge_base_id #=> String
event.trace.orchestration_trace.invocation_input.knowledge_base_lookup_input.text #=> String
event.trace.orchestration_trace.invocation_input.trace_id #=> String
event.trace.orchestration_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.orchestration_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.orchestration_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.orchestration_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.orchestration_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.orchestration_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.orchestration_trace.model_invocation_input.override_lambda #=> String
event.trace.orchestration_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.orchestration_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.orchestration_trace.model_invocation_input.text #=> String
event.trace.orchestration_trace.model_invocation_input.trace_id #=> String
event.trace.orchestration_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.orchestration_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.orchestration_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.orchestration_trace.model_invocation_output.raw_response.content #=> String
event.trace.orchestration_trace.model_invocation_output.trace_id #=> String
event.trace.orchestration_trace.observation.action_group_invocation_output.text #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_error #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_output #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_timeout #=> Boolean
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.files #=> Array
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.files[0] #=> String
event.trace.orchestration_trace.observation.final_response.text #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references #=> Array
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].content.text #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.confluence_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.s3_location.uri #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.salesforce_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.share_point_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.web_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].metadata #=> Hash
event.trace.orchestration_trace.observation.reprompt_response.source #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "PARSER"
event.trace.orchestration_trace.observation.reprompt_response.text #=> String
event.trace.orchestration_trace.observation.trace_id #=> String
event.trace.orchestration_trace.observation.type #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "FINISH", "ASK_USER", "REPROMPT"
event.trace.orchestration_trace.rationale.text #=> String
event.trace.orchestration_trace.rationale.trace_id #=> String
event.trace.post_processing_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.post_processing_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.post_processing_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.post_processing_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.post_processing_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.post_processing_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.post_processing_trace.model_invocation_input.override_lambda #=> String
event.trace.post_processing_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.post_processing_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.post_processing_trace.model_invocation_input.text #=> String
event.trace.post_processing_trace.model_invocation_input.trace_id #=> String
event.trace.post_processing_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.post_processing_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.post_processing_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.post_processing_trace.model_invocation_output.parsed_response.text #=> String
event.trace.post_processing_trace.model_invocation_output.raw_response.content #=> String
event.trace.post_processing_trace.model_invocation_output.trace_id #=> String
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.pre_processing_trace.model_invocation_input.override_lambda #=> String
event.trace.pre_processing_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.pre_processing_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.pre_processing_trace.model_invocation_input.text #=> String
event.trace.pre_processing_trace.model_invocation_input.trace_id #=> String
event.trace.pre_processing_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.pre_processing_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.pre_processing_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.pre_processing_trace.model_invocation_output.parsed_response.is_valid #=> Boolean
event.trace.pre_processing_trace.model_invocation_output.parsed_response.rationale #=> String
event.trace.pre_processing_trace.model_invocation_output.raw_response.content #=> String
event.trace.pre_processing_trace.model_invocation_output.trace_id #=> String
For :validation_exception event available at #on_validation_exception_event callback and response eventstream enumerator:
event.message #=> String
resp.content_type #=> String
resp.memory_id #=> String
resp.session_id #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:agent_alias_id
(required, String)
—
The alias of the agent to use.
-
:agent_id
(required, String)
—
The unique identifier of the agent to use.
-
:enable_trace
(Boolean)
—
Specifies whether to turn on the trace or not to track the agent's reasoning process. For more information, see Trace enablement.
-
:end_session
(Boolean)
—
Specifies whether to end the session with the agent or not.
-
:input_text
(String)
—
The prompt text to send the agent.
If you include returnControlInvocationResults
in thesessionState
field, theinputText
field will be ignored. -
:memory_id
(String)
—
The unique identifier of the agent memory.
-
:session_id
(required, String)
—
The unique identifier of the session. Use the same value across requests to continue the same conversation.
-
:session_state
(Types::SessionState)
—
Contains parameters that specify various attributes of the session. For more information, see Control session context.
If you include returnControlInvocationResults
in thesessionState
field, theinputText
field will be ignored.
Yields:
- (event_stream_handler)
Returns:
-
(Types::InvokeAgentResponse)
—
Returns a response object which responds to the following methods:
- #completion => Types::ResponseStream
- #content_type => String
- #memory_id => String
- #session_id => String
See Also:
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 1225 def invoke_agent(params = {}, options = {}, &block) params = params.dup event_stream_handler = case handler = params.delete(:event_stream_handler) when EventStreams::ResponseStream then handler when Proc then EventStreams::ResponseStream.new.tap(&handler) when nil then EventStreams::ResponseStream.new else msg = "expected :event_stream_handler to be a block or "\ "instance of Aws::BedrockAgentRuntime::EventStreams::ResponseStream"\ ", got `#{handler.inspect}` instead" raise ArgumentError, msg end yield(event_stream_handler) if block_given? req = build_request(:invoke_agent, params) req.context[:event_stream_handler] = event_stream_handler req.handlers.add(Aws::Binary::DecodeHandler, priority: 95) req.send_request(options, &block) end |
#invoke_flow(params = {}) ⇒ Types::InvokeFlowResponse
Invokes an alias of a flow to run the inputs that you specify and return the output of each node as a stream. If there's an error, the error is returned. For more information, see Test a flow in Amazon Bedrock in the Amazon Bedrock User Guide.
InvokeFlow
.
Examples:
EventStream Operation Example
EventStream Operation Example
You can process the event once it arrives immediately, or wait until the
full response is complete and iterate through the eventstream enumerator.
To interact with event immediately, you need to register #invoke_flow
with callbacks. Callbacks can be registered for specific events or for all
events, including error events.
Callbacks can be passed into the `:event_stream_handler` option or within a
block statement attached to the #invoke_flow call directly. Hybrid
pattern of both is also supported.
`:event_stream_handler` option takes in either a Proc object or
Aws::BedrockAgentRuntime::EventStreams::FlowResponseStream object.
Usage pattern a): Callbacks with a block attached to #invoke_flow
Example for registering callbacks for all event types and an error event
client.invoke_flow( # params input# ) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
stream.on_event do |event|
# process all events arrive
puts event.event_type
...
end
end
Usage pattern b): Pass in `:event_stream_handler` for #invoke_flow
1) Create a Aws::BedrockAgentRuntime::EventStreams::FlowResponseStream object
Example for registering callbacks with specific events
handler = Aws::BedrockAgentRuntime::EventStreams::FlowResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_flow_completion_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowCompletionEvent
end
handler.on_flow_output_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowOutputEvent
end
handler.on_flow_trace_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowTraceEvent
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_flow( # params input #, event_stream_handler: handler)
2) Use a Ruby Proc object
Example for registering callbacks with specific events
handler = Proc.new do |stream|
stream.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
stream.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
stream.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
stream.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
stream.on_flow_completion_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowCompletionEvent
end
stream.on_flow_output_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowOutputEvent
end
stream.on_flow_trace_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowTraceEvent
end
stream.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
stream.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
stream.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
stream.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
stream.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
end
client.invoke_flow( # params input #, event_stream_handler: handler)
Usage pattern c): Hybrid pattern of a) and b)
handler = Aws::BedrockAgentRuntime::EventStreams::FlowResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_flow_completion_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowCompletionEvent
end
handler.on_flow_output_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowOutputEvent
end
handler.on_flow_trace_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::flowTraceEvent
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_flow( # params input #, event_stream_handler: handler) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
end
You can also iterate through events after the response complete.
Events are available at resp.response_stream # => Enumerator
For parameter input example, please refer to following request syntax
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.invoke_flow({
enable_trace: false,
flow_alias_identifier: "FlowAliasIdentifier", # required
flow_identifier: "FlowIdentifier", # required
inputs: [ # required
{
content: { # required
document: {
},
},
node_name: "NodeName", # required
node_output_name: "NodeOutputName", # required
},
],
})
Response structure
Response structure
All events are available at resp.response_stream:
resp.response_stream #=> Enumerator
resp.response_stream.event_types #=> [:access_denied_exception, :bad_gateway_exception, :conflict_exception, :dependency_failed_exception, :flow_completion_event, :flow_output_event, :flow_trace_event, :internal_server_exception, :resource_not_found_exception, :service_quota_exceeded_exception, :throttling_exception, :validation_exception]
For :access_denied_exception event available at #on_access_denied_exception_event callback and response eventstream enumerator:
event.message #=> String
For :bad_gateway_exception event available at #on_bad_gateway_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :conflict_exception event available at #on_conflict_exception_event callback and response eventstream enumerator:
event.message #=> String
For :dependency_failed_exception event available at #on_dependency_failed_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :flow_completion_event event available at #on_flow_completion_event_event callback and response eventstream enumerator:
event.completion_reason #=> String, one of "SUCCESS"
For :flow_output_event event available at #on_flow_output_event_event callback and response eventstream enumerator:
event.node_name #=> String
event.node_type #=> String, one of "FlowInputNode", "FlowOutputNode", "LambdaFunctionNode", "KnowledgeBaseNode", "PromptNode", "ConditionNode", "LexNode"
For :flow_trace_event event available at #on_flow_trace_event_event callback and response eventstream enumerator:
event.trace.condition_node_result_trace.node_name #=> String
event.trace.condition_node_result_trace.satisfied_conditions #=> Array
event.trace.condition_node_result_trace.satisfied_conditions[0].condition_name #=> String
event.trace.condition_node_result_trace.timestamp #=> Time
event.trace.node_input_trace.fields #=> Array
event.trace.node_input_trace.fields[0].node_input_name #=> String
event.trace.node_input_trace.node_name #=> String
event.trace.node_input_trace.timestamp #=> Time
event.trace.node_output_trace.fields #=> Array
event.trace.node_output_trace.fields[0].node_output_name #=> String
event.trace.node_output_trace.node_name #=> String
event.trace.node_output_trace.timestamp #=> Time
For :internal_server_exception event available at #on_internal_server_exception_event callback and response eventstream enumerator:
event.message #=> String
For :resource_not_found_exception event available at #on_resource_not_found_exception_event callback and response eventstream enumerator:
event.message #=> String
For :service_quota_exceeded_exception event available at #on_service_quota_exceeded_exception_event callback and response eventstream enumerator:
event.message #=> String
For :throttling_exception event available at #on_throttling_exception_event callback and response eventstream enumerator:
event.message #=> String
For :validation_exception event available at #on_validation_exception_event callback and response eventstream enumerator:
event.message #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:enable_trace
(Boolean)
—
Specifies whether to return the trace for the flow or not. Traces track inputs and outputs for nodes in the flow. For more information, see Track each step in your prompt flow by viewing its trace in Amazon Bedrock.
-
:flow_alias_identifier
(required, String)
—
The unique identifier of the flow alias.
-
:flow_identifier
(required, String)
—
The unique identifier of the flow.
-
:inputs
(required, Array<Types::FlowInput>)
—
A list of objects, each containing information about an input into the flow.
Yields:
- (event_stream_handler)
Returns:
-
(Types::InvokeFlowResponse)
—
Returns a response object which responds to the following methods:
- #response_stream => Types::FlowResponseStream
See Also:
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 1547 def invoke_flow(params = {}, options = {}, &block) params = params.dup event_stream_handler = case handler = params.delete(:event_stream_handler) when EventStreams::FlowResponseStream then handler when Proc then EventStreams::FlowResponseStream.new.tap(&handler) when nil then EventStreams::FlowResponseStream.new else msg = "expected :event_stream_handler to be a block or "\ "instance of Aws::BedrockAgentRuntime::EventStreams::FlowResponseStream"\ ", got `#{handler.inspect}` instead" raise ArgumentError, msg end yield(event_stream_handler) if block_given? req = build_request(:invoke_flow, params) req.context[:event_stream_handler] = event_stream_handler req.handlers.add(Aws::Binary::DecodeHandler, priority: 95) req.send_request(options, &block) end |
#invoke_inline_agent(params = {}) ⇒ Types::InvokeInlineAgentResponse
Invokes an inline Amazon Bedrock agent using the configurations you provide with the request.
Specify the following fields for security purposes.
(Optional)
customerEncryptionKeyArn
– The Amazon Resource Name (ARN) of a KMS key to encrypt the creation of the agent.(Optional)
idleSessionTTLinSeconds
– Specify the number of seconds for which the agent should maintain session information. After this time expires, the subsequentInvokeInlineAgent
request begins a new session.
To override the default prompt behavior for agent orchestration and to use advanced prompts, include a
promptOverrideConfiguration
object. For more information, see Advanced prompts.The agent instructions will not be honored if your agent has only one knowledge base, uses default prompts, has no action group, and user input is disabled.
InvokeInlineAgent
.
Examples:
EventStream Operation Example
EventStream Operation Example
You can process the event once it arrives immediately, or wait until the
full response is complete and iterate through the eventstream enumerator.
To interact with event immediately, you need to register #invoke_inline_agent
with callbacks. Callbacks can be registered for specific events or for all
events, including error events.
Callbacks can be passed into the `:event_stream_handler` option or within a
block statement attached to the #invoke_inline_agent call directly. Hybrid
pattern of both is also supported.
`:event_stream_handler` option takes in either a Proc object or
Aws::BedrockAgentRuntime::EventStreams::InlineAgentResponseStream object.
Usage pattern a): Callbacks with a block attached to #invoke_inline_agent
Example for registering callbacks for all event types and an error event
client.invoke_inline_agent( # params input# ) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
stream.on_event do |event|
# process all events arrive
puts event.event_type
...
end
end
Usage pattern b): Pass in `:event_stream_handler` for #invoke_inline_agent
1) Create a Aws::BedrockAgentRuntime::EventStreams::InlineAgentResponseStream object
Example for registering callbacks with specific events
handler = Aws::BedrockAgentRuntime::EventStreams::InlineAgentResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_inline_agent( # params input #, event_stream_handler: handler)
2) Use a Ruby Proc object
Example for registering callbacks with specific events
handler = Proc.new do |stream|
stream.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
stream.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
stream.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
stream.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
stream.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
stream.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
stream.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
stream.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
stream.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
stream.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
stream.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
stream.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
stream.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
end
client.invoke_inline_agent( # params input #, event_stream_handler: handler)
Usage pattern c): Hybrid pattern of a) and b)
handler = Aws::BedrockAgentRuntime::EventStreams::InlineAgentResponseStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_chunk_event do |event|
event # => Aws::BedrockAgentRuntime::Types::chunk
end
handler.on_conflict_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::conflictException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_files_event do |event|
event # => Aws::BedrockAgentRuntime::Types::files
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_resource_not_found_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::resourceNotFoundException
end
handler.on_return_control_event do |event|
event # => Aws::BedrockAgentRuntime::Types::returnControl
end
handler.on_service_quota_exceeded_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::serviceQuotaExceededException
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_trace_event do |event|
event # => Aws::BedrockAgentRuntime::Types::trace
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.invoke_inline_agent( # params input #, event_stream_handler: handler) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
end
You can also iterate through events after the response complete.
Events are available at resp.completion # => Enumerator
For parameter input example, please refer to following request syntax
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.invoke_inline_agent({
action_groups: [
{
action_group_executor: {
custom_control: "RETURN_CONTROL", # accepts RETURN_CONTROL
lambda: "LambdaResourceArn",
},
action_group_name: "ResourceName", # required
api_schema: {
payload: "Payload",
s3: {
s3_bucket_name: "S3BucketName",
s3_object_key: "S3ObjectKey",
},
},
description: "ResourceDescription",
function_schema: {
functions: [
{
description: "FunctionDescription",
name: "ResourceName", # required
parameters: {
"ParameterName" => {
description: "ParameterDescription",
required: false,
type: "string", # required, accepts string, number, integer, boolean, array
},
},
require_confirmation: "ENABLED", # accepts ENABLED, DISABLED
},
],
},
parent_action_group_signature: "AMAZON.UserInput", # accepts AMAZON.UserInput, AMAZON.CodeInterpreter
},
],
customer_encryption_key_arn: "KmsKeyArn",
enable_trace: false,
end_session: false,
foundation_model: "ModelIdentifier", # required
guardrail_configuration: {
guardrail_identifier: "GuardrailIdentifierWithArn", # required
guardrail_version: "GuardrailVersion", # required
},
idle_session_ttl_in_seconds: 1,
inline_session_state: {
files: [
{
name: "String", # required
source: { # required
byte_content: {
data: "data", # required
media_type: "MimeType", # required
},
s3_location: {
uri: "S3Uri", # required
},
source_type: "S3", # required, accepts S3, BYTE_CONTENT
},
use_case: "CODE_INTERPRETER", # required, accepts CODE_INTERPRETER, CHAT
},
],
invocation_id: "String",
prompt_session_attributes: {
"String" => "String",
},
return_control_invocation_results: [
{
api_result: {
action_group: "String", # required
api_path: "ApiPath",
confirmation_state: "CONFIRM", # accepts CONFIRM, DENY
http_method: "String",
http_status_code: 1,
response_body: {
"String" => {
body: "String",
},
},
response_state: "FAILURE", # accepts FAILURE, REPROMPT
},
function_result: {
action_group: "String", # required
confirmation_state: "CONFIRM", # accepts CONFIRM, DENY
function: "String",
response_body: {
"String" => {
body: "String",
},
},
response_state: "FAILURE", # accepts FAILURE, REPROMPT
},
},
],
session_attributes: {
"String" => "String",
},
},
input_text: "InputText",
instruction: "Instruction", # required
knowledge_bases: [
{
description: "ResourceDescription", # required
knowledge_base_id: "KnowledgeBaseId", # required
retrieval_configuration: {
vector_search_configuration: { # required
filter: {
and_all: [
{
# recursive RetrievalFilter
},
],
equals: {
key: "FilterKey", # required
value: { # required
},
},
greater_than: {
key: "FilterKey", # required
value: { # required
},
},
greater_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
in: {
key: "FilterKey", # required
value: { # required
},
},
less_than: {
key: "FilterKey", # required
value: { # required
},
},
less_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
list_contains: {
key: "FilterKey", # required
value: { # required
},
},
not_equals: {
key: "FilterKey", # required
value: { # required
},
},
not_in: {
key: "FilterKey", # required
value: { # required
},
},
or_all: [
{
# recursive RetrievalFilter
},
],
starts_with: {
key: "FilterKey", # required
value: { # required
},
},
string_contains: {
key: "FilterKey", # required
value: { # required
},
},
},
number_of_results: 1,
override_search_type: "HYBRID", # accepts HYBRID, SEMANTIC
},
},
},
],
prompt_override_configuration: {
override_lambda: "LambdaResourceArn",
prompt_configurations: [ # required
{
base_prompt_template: "BasePromptTemplate",
inference_configuration: {
maximum_length: 1,
stop_sequences: ["String"],
temperature: 1.0,
top_k: 1,
top_p: 1.0,
},
parser_mode: "DEFAULT", # accepts DEFAULT, OVERRIDDEN
prompt_creation_mode: "DEFAULT", # accepts DEFAULT, OVERRIDDEN
prompt_state: "ENABLED", # accepts ENABLED, DISABLED
prompt_type: "PRE_PROCESSING", # accepts PRE_PROCESSING, ORCHESTRATION, KNOWLEDGE_BASE_RESPONSE_GENERATION, POST_PROCESSING, ROUTING_CLASSIFIER
},
],
},
session_id: "SessionId", # required
})
Response structure
Response structure
All events are available at resp.completion:
resp.completion #=> Enumerator
resp.completion.event_types #=> [:access_denied_exception, :bad_gateway_exception, :chunk, :conflict_exception, :dependency_failed_exception, :files, :internal_server_exception, :resource_not_found_exception, :return_control, :service_quota_exceeded_exception, :throttling_exception, :trace, :validation_exception]
For :access_denied_exception event available at #on_access_denied_exception_event callback and response eventstream enumerator:
event.message #=> String
For :bad_gateway_exception event available at #on_bad_gateway_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :chunk event available at #on_chunk_event callback and response eventstream enumerator:
event.attribution.citations #=> Array
event.attribution.citations[0].generated_response_part.text_response_part.span.end #=> Integer
event.attribution.citations[0].generated_response_part.text_response_part.span.start #=> Integer
event.attribution.citations[0].generated_response_part.text_response_part.text #=> String
event.attribution.citations[0].retrieved_references #=> Array
event.attribution.citations[0].retrieved_references[0].content.text #=> String
event.attribution.citations[0].retrieved_references[0].location.confluence_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.s3_location.uri #=> String
event.attribution.citations[0].retrieved_references[0].location.salesforce_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.share_point_location.url #=> String
event.attribution.citations[0].retrieved_references[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
event.attribution.citations[0].retrieved_references[0].location.web_location.url #=> String
event.attribution.citations[0].retrieved_references[0].metadata #=> Hash
event.bytes #=> String
For :conflict_exception event available at #on_conflict_exception_event callback and response eventstream enumerator:
event.message #=> String
For :dependency_failed_exception event available at #on_dependency_failed_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :files event available at #on_files_event callback and response eventstream enumerator:
event.files #=> Array
event.files[0].bytes #=> String
event.files[0].name #=> String
event.files[0].type #=> String
For :internal_server_exception event available at #on_internal_server_exception_event callback and response eventstream enumerator:
event.message #=> String
For :resource_not_found_exception event available at #on_resource_not_found_exception_event callback and response eventstream enumerator:
event.message #=> String
For :return_control event available at #on_return_control_event callback and response eventstream enumerator:
event.invocation_id #=> String
event.invocation_inputs #=> Array
event.invocation_inputs[0].api_invocation_input.action_group #=> String
event.invocation_inputs[0].api_invocation_input.action_invocation_type #=> String, one of "RESULT", "USER_CONFIRMATION", "USER_CONFIRMATION_AND_RESULT"
event.invocation_inputs[0].api_invocation_input.api_path #=> String
event.invocation_inputs[0].api_invocation_input.http_method #=> String
event.invocation_inputs[0].api_invocation_input.parameters #=> Array
event.invocation_inputs[0].api_invocation_input.parameters[0].name #=> String
event.invocation_inputs[0].api_invocation_input.parameters[0].type #=> String
event.invocation_inputs[0].api_invocation_input.parameters[0].value #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content #=> Hash
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties #=> Array
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].name #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].type #=> String
event.invocation_inputs[0].api_invocation_input.request_body.content["String"].properties[0].value #=> String
event.invocation_inputs[0].function_invocation_input.action_group #=> String
event.invocation_inputs[0].function_invocation_input.action_invocation_type #=> String, one of "RESULT", "USER_CONFIRMATION", "USER_CONFIRMATION_AND_RESULT"
event.invocation_inputs[0].function_invocation_input.function #=> String
event.invocation_inputs[0].function_invocation_input.parameters #=> Array
event.invocation_inputs[0].function_invocation_input.parameters[0].name #=> String
event.invocation_inputs[0].function_invocation_input.parameters[0].type #=> String
event.invocation_inputs[0].function_invocation_input.parameters[0].value #=> String
For :service_quota_exceeded_exception event available at #on_service_quota_exceeded_exception_event callback and response eventstream enumerator:
event.message #=> String
For :throttling_exception event available at #on_throttling_exception_event callback and response eventstream enumerator:
event.message #=> String
For :trace event available at #on_trace_event callback and response eventstream enumerator:
event.session_id #=> String
event.trace.failure_trace.failure_reason #=> String
event.trace.failure_trace.trace_id #=> String
event.trace.guardrail_trace.action #=> String, one of "INTERVENED", "NONE"
event.trace.guardrail_trace.input_assessments #=> Array
event.trace.guardrail_trace.input_assessments[0].content_policy.filters #=> Array
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].confidence #=> String, one of "NONE", "LOW", "MEDIUM", "HIGH"
event.trace.guardrail_trace.input_assessments[0].content_policy.filters[0].type #=> String, one of "INSULTS", "HATE", "SEXUAL", "VIOLENCE", "MISCONDUCT", "PROMPT_ATTACK"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities #=> Array
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.pii_entities[0].type #=> String, one of "ADDRESS", "AGE", "AWS_ACCESS_KEY", "AWS_SECRET_KEY", "CA_HEALTH_NUMBER", "CA_SOCIAL_INSURANCE_NUMBER", "CREDIT_DEBIT_CARD_CVV", "CREDIT_DEBIT_CARD_EXPIRY", "CREDIT_DEBIT_CARD_NUMBER", "DRIVER_ID", "EMAIL", "INTERNATIONAL_BANK_ACCOUNT_NUMBER", "IP_ADDRESS", "LICENSE_PLATE", "MAC_ADDRESS", "NAME", "PASSWORD", "PHONE", "PIN", "SWIFT_CODE", "UK_NATIONAL_HEALTH_SERVICE_NUMBER", "UK_NATIONAL_INSURANCE_NUMBER", "UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER", "URL", "USERNAME", "US_BANK_ACCOUNT_NUMBER", "US_BANK_ROUTING_NUMBER", "US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER", "US_PASSPORT_NUMBER", "US_SOCIAL_SECURITY_NUMBER", "VEHICLE_IDENTIFICATION_NUMBER"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes #=> Array
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].name #=> String
event.trace.guardrail_trace.input_assessments[0].sensitive_information_policy.regexes[0].regex #=> String
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics #=> Array
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].name #=> String
event.trace.guardrail_trace.input_assessments[0].topic_policy.topics[0].type #=> String, one of "DENY"
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words #=> Array
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].word_policy.custom_words[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists #=> Array
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].match #=> String
event.trace.guardrail_trace.input_assessments[0].word_policy.managed_word_lists[0].type #=> String, one of "PROFANITY"
event.trace.guardrail_trace.output_assessments #=> Array
event.trace.guardrail_trace.output_assessments[0].content_policy.filters #=> Array
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].confidence #=> String, one of "NONE", "LOW", "MEDIUM", "HIGH"
event.trace.guardrail_trace.output_assessments[0].content_policy.filters[0].type #=> String, one of "INSULTS", "HATE", "SEXUAL", "VIOLENCE", "MISCONDUCT", "PROMPT_ATTACK"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities #=> Array
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.pii_entities[0].type #=> String, one of "ADDRESS", "AGE", "AWS_ACCESS_KEY", "AWS_SECRET_KEY", "CA_HEALTH_NUMBER", "CA_SOCIAL_INSURANCE_NUMBER", "CREDIT_DEBIT_CARD_CVV", "CREDIT_DEBIT_CARD_EXPIRY", "CREDIT_DEBIT_CARD_NUMBER", "DRIVER_ID", "EMAIL", "INTERNATIONAL_BANK_ACCOUNT_NUMBER", "IP_ADDRESS", "LICENSE_PLATE", "MAC_ADDRESS", "NAME", "PASSWORD", "PHONE", "PIN", "SWIFT_CODE", "UK_NATIONAL_HEALTH_SERVICE_NUMBER", "UK_NATIONAL_INSURANCE_NUMBER", "UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER", "URL", "USERNAME", "US_BANK_ACCOUNT_NUMBER", "US_BANK_ROUTING_NUMBER", "US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER", "US_PASSPORT_NUMBER", "US_SOCIAL_SECURITY_NUMBER", "VEHICLE_IDENTIFICATION_NUMBER"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes #=> Array
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].action #=> String, one of "BLOCKED", "ANONYMIZED"
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].name #=> String
event.trace.guardrail_trace.output_assessments[0].sensitive_information_policy.regexes[0].regex #=> String
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics #=> Array
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].name #=> String
event.trace.guardrail_trace.output_assessments[0].topic_policy.topics[0].type #=> String, one of "DENY"
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words #=> Array
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].word_policy.custom_words[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists #=> Array
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].action #=> String, one of "BLOCKED"
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].match #=> String
event.trace.guardrail_trace.output_assessments[0].word_policy.managed_word_lists[0].type #=> String, one of "PROFANITY"
event.trace.guardrail_trace.trace_id #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.action_group_name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.api_path #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.execution_type #=> String, one of "LAMBDA", "RETURN_CONTROL"
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.function #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.invocation_id #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters #=> Array
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].type #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.parameters[0].value #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content #=> Hash
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"] #=> Array
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].name #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].type #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.request_body.content["String"][0].value #=> String
event.trace.orchestration_trace.invocation_input.action_group_invocation_input.verb #=> String
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.code #=> String
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.files #=> Array
event.trace.orchestration_trace.invocation_input.code_interpreter_invocation_input.files[0] #=> String
event.trace.orchestration_trace.invocation_input.invocation_type #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "FINISH", "ACTION_GROUP_CODE_INTERPRETER"
event.trace.orchestration_trace.invocation_input.knowledge_base_lookup_input.knowledge_base_id #=> String
event.trace.orchestration_trace.invocation_input.knowledge_base_lookup_input.text #=> String
event.trace.orchestration_trace.invocation_input.trace_id #=> String
event.trace.orchestration_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.orchestration_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.orchestration_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.orchestration_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.orchestration_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.orchestration_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.orchestration_trace.model_invocation_input.override_lambda #=> String
event.trace.orchestration_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.orchestration_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.orchestration_trace.model_invocation_input.text #=> String
event.trace.orchestration_trace.model_invocation_input.trace_id #=> String
event.trace.orchestration_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.orchestration_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.orchestration_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.orchestration_trace.model_invocation_output.raw_response.content #=> String
event.trace.orchestration_trace.model_invocation_output.trace_id #=> String
event.trace.orchestration_trace.observation.action_group_invocation_output.text #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_error #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_output #=> String
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.execution_timeout #=> Boolean
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.files #=> Array
event.trace.orchestration_trace.observation.code_interpreter_invocation_output.files[0] #=> String
event.trace.orchestration_trace.observation.final_response.text #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references #=> Array
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].content.text #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.confluence_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.s3_location.uri #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.salesforce_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.share_point_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].location.web_location.url #=> String
event.trace.orchestration_trace.observation.knowledge_base_lookup_output.retrieved_references[0].metadata #=> Hash
event.trace.orchestration_trace.observation.reprompt_response.source #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "PARSER"
event.trace.orchestration_trace.observation.reprompt_response.text #=> String
event.trace.orchestration_trace.observation.trace_id #=> String
event.trace.orchestration_trace.observation.type #=> String, one of "ACTION_GROUP", "KNOWLEDGE_BASE", "FINISH", "ASK_USER", "REPROMPT"
event.trace.orchestration_trace.rationale.text #=> String
event.trace.orchestration_trace.rationale.trace_id #=> String
event.trace.post_processing_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.post_processing_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.post_processing_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.post_processing_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.post_processing_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.post_processing_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.post_processing_trace.model_invocation_input.override_lambda #=> String
event.trace.post_processing_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.post_processing_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.post_processing_trace.model_invocation_input.text #=> String
event.trace.post_processing_trace.model_invocation_input.trace_id #=> String
event.trace.post_processing_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.post_processing_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.post_processing_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.post_processing_trace.model_invocation_output.parsed_response.text #=> String
event.trace.post_processing_trace.model_invocation_output.raw_response.content #=> String
event.trace.post_processing_trace.model_invocation_output.trace_id #=> String
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.maximum_length #=> Integer
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.stop_sequences #=> Array
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.stop_sequences[0] #=> String
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.temperature #=> Float
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.top_k #=> Integer
event.trace.pre_processing_trace.model_invocation_input.inference_configuration.top_p #=> Float
event.trace.pre_processing_trace.model_invocation_input.override_lambda #=> String
event.trace.pre_processing_trace.model_invocation_input.parser_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.pre_processing_trace.model_invocation_input.prompt_creation_mode #=> String, one of "DEFAULT", "OVERRIDDEN"
event.trace.pre_processing_trace.model_invocation_input.text #=> String
event.trace.pre_processing_trace.model_invocation_input.trace_id #=> String
event.trace.pre_processing_trace.model_invocation_input.type #=> String, one of "PRE_PROCESSING", "ORCHESTRATION", "KNOWLEDGE_BASE_RESPONSE_GENERATION", "POST_PROCESSING", "ROUTING_CLASSIFIER"
event.trace.pre_processing_trace.model_invocation_output.metadata.usage.input_tokens #=> Integer
event.trace.pre_processing_trace.model_invocation_output.metadata.usage.output_tokens #=> Integer
event.trace.pre_processing_trace.model_invocation_output.parsed_response.is_valid #=> Boolean
event.trace.pre_processing_trace.model_invocation_output.parsed_response.rationale #=> String
event.trace.pre_processing_trace.model_invocation_output.raw_response.content #=> String
event.trace.pre_processing_trace.model_invocation_output.trace_id #=> String
For :validation_exception event available at #on_validation_exception_event callback and response eventstream enumerator:
event.message #=> String
resp.content_type #=> String
resp.session_id #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:action_groups
(Array<Types::AgentActionGroup>)
—
A list of action groups with each action group defining the action the inline agent needs to carry out.
-
:customer_encryption_key_arn
(String)
—
The Amazon Resource Name (ARN) of the Amazon Web Services KMS key to use to encrypt your inline agent.
-
:enable_trace
(Boolean)
—
Specifies whether to turn on the trace or not to track the agent's reasoning process. For more information, see Using trace.
-
:end_session
(Boolean)
—
Specifies whether to end the session with the inline agent or not.
-
:foundation_model
(required, String)
—
The model identifier (ID) of the model to use for orchestration by the inline agent. For example,
meta.llama3-1-70b-instruct-v1:0
. -
:guardrail_configuration
(Types::GuardrailConfigurationWithArn)
—
The guardrails to assign to the inline agent.
-
:idle_session_ttl_in_seconds
(Integer)
—
The number of seconds for which the inline agent should maintain session information. After this time expires, the subsequent
InvokeInlineAgent
request begins a new session.A user interaction remains active for the amount of time specified. If no conversation occurs during this time, the session expires and the data provided before the timeout is deleted.
-
:inline_session_state
(Types::InlineSessionState)
—
Parameters that specify the various attributes of a sessions. You can include attributes for the session or prompt or, if you configured an action group to return control, results from invocation of the action group. For more information, see Control session context.
If you include returnControlInvocationResults
in thesessionState
field, theinputText
field will be ignored. -
:input_text
(String)
—
The prompt text to send to the agent.
If you include returnControlInvocationResults
in thesessionState
field, theinputText
field will be ignored. -
:instruction
(required, String)
—
The instructions that tell the inline agent what it should do and how it should interact with users.
-
:knowledge_bases
(Array<Types::KnowledgeBase>)
—
Contains information of the knowledge bases to associate with.
-
:prompt_override_configuration
(Types::PromptOverrideConfiguration)
—
Configurations for advanced prompts used to override the default prompts to enhance the accuracy of the inline agent.
-
:session_id
(required, String)
—
The unique identifier of the session. Use the same value across requests to continue the same conversation.
Yields:
- (event_stream_handler)
Returns:
-
(Types::InvokeInlineAgentResponse)
—
Returns a response object which responds to the following methods:
- #completion => Types::InlineAgentResponseStream
- #content_type => String
- #session_id => String
See Also:
2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 2320 def invoke_inline_agent(params = {}, options = {}, &block) params = params.dup event_stream_handler = case handler = params.delete(:event_stream_handler) when EventStreams::InlineAgentResponseStream then handler when Proc then EventStreams::InlineAgentResponseStream.new.tap(&handler) when nil then EventStreams::InlineAgentResponseStream.new else msg = "expected :event_stream_handler to be a block or "\ "instance of Aws::BedrockAgentRuntime::EventStreams::InlineAgentResponseStream"\ ", got `#{handler.inspect}` instead" raise ArgumentError, msg end yield(event_stream_handler) if block_given? req = build_request(:invoke_inline_agent, params) req.context[:event_stream_handler] = event_stream_handler req.handlers.add(Aws::Binary::DecodeHandler, priority: 95) req.send_request(options, &block) end |
#optimize_prompt(params = {}) ⇒ Types::OptimizePromptResponse
Optimizes a prompt for the task that you specify. For more information, see Optimize a prompt in the Amazon Bedrock User Guide.
Examples:
EventStream Operation Example
EventStream Operation Example
You can process the event once it arrives immediately, or wait until the
full response is complete and iterate through the eventstream enumerator.
To interact with event immediately, you need to register #optimize_prompt
with callbacks. Callbacks can be registered for specific events or for all
events, including error events.
Callbacks can be passed into the `:event_stream_handler` option or within a
block statement attached to the #optimize_prompt call directly. Hybrid
pattern of both is also supported.
`:event_stream_handler` option takes in either a Proc object or
Aws::BedrockAgentRuntime::EventStreams::OptimizedPromptStream object.
Usage pattern a): Callbacks with a block attached to #optimize_prompt
Example for registering callbacks for all event types and an error event
client.optimize_prompt( # params input# ) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
stream.on_event do |event|
# process all events arrive
puts event.event_type
...
end
end
Usage pattern b): Pass in `:event_stream_handler` for #optimize_prompt
1) Create a Aws::BedrockAgentRuntime::EventStreams::OptimizedPromptStream object
Example for registering callbacks with specific events
handler = Aws::BedrockAgentRuntime::EventStreams::OptimizedPromptStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_analyze_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::analyzePromptEvent
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_optimized_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::optimizedPromptEvent
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.optimize_prompt( # params input #, event_stream_handler: handler)
2) Use a Ruby Proc object
Example for registering callbacks with specific events
handler = Proc.new do |stream|
stream.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
stream.on_analyze_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::analyzePromptEvent
end
stream.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
stream.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
stream.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
stream.on_optimized_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::optimizedPromptEvent
end
stream.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
stream.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
end
client.optimize_prompt( # params input #, event_stream_handler: handler)
Usage pattern c): Hybrid pattern of a) and b)
handler = Aws::BedrockAgentRuntime::EventStreams::OptimizedPromptStream.new
handler.on_access_denied_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::accessDeniedException
end
handler.on_analyze_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::analyzePromptEvent
end
handler.on_bad_gateway_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::badGatewayException
end
handler.on_dependency_failed_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::dependencyFailedException
end
handler.on_internal_server_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::internalServerException
end
handler.on_optimized_prompt_event_event do |event|
event # => Aws::BedrockAgentRuntime::Types::optimizedPromptEvent
end
handler.on_throttling_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::throttlingException
end
handler.on_validation_exception_event do |event|
event # => Aws::BedrockAgentRuntime::Types::validationException
end
client.optimize_prompt( # params input #, event_stream_handler: handler) do |stream|
stream.on_error_event do |event|
# catch unmodeled error event in the stream
raise event
# => Aws::Errors::EventError
# event.event_type => :error
# event.error_code => String
# event.error_message => String
end
end
You can also iterate through events after the response complete.
Events are available at resp.optimized_prompt # => Enumerator
For parameter input example, please refer to following request syntax
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.optimize_prompt({
input: { # required
text_prompt: {
text: "TextPromptTextString", # required
},
},
target_model_id: "OptimizePromptRequestTargetModelIdString", # required
})
Response structure
Response structure
All events are available at resp.optimized_prompt:
resp.optimized_prompt #=> Enumerator
resp.optimized_prompt.event_types #=> [:access_denied_exception, :analyze_prompt_event, :bad_gateway_exception, :dependency_failed_exception, :internal_server_exception, :optimized_prompt_event, :throttling_exception, :validation_exception]
For :access_denied_exception event available at #on_access_denied_exception_event callback and response eventstream enumerator:
event.message #=> String
For :analyze_prompt_event event available at #on_analyze_prompt_event_event callback and response eventstream enumerator:
event.message #=> String
For :bad_gateway_exception event available at #on_bad_gateway_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :dependency_failed_exception event available at #on_dependency_failed_exception_event callback and response eventstream enumerator:
event.message #=> String
event.resource_name #=> String
For :internal_server_exception event available at #on_internal_server_exception_event callback and response eventstream enumerator:
event.message #=> String
For :optimized_prompt_event event available at #on_optimized_prompt_event_event callback and response eventstream enumerator:
event.optimized_prompt.text_prompt.text #=> String
For :throttling_exception event available at #on_throttling_exception_event callback and response eventstream enumerator:
event.message #=> String
For :validation_exception event available at #on_validation_exception_event callback and response eventstream enumerator:
event.message #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:input
(required, Types::InputPrompt)
—
Contains the prompt to optimize.
-
:target_model_id
(required, String)
—
The unique identifier of the model that you want to optimize the prompt for.
Yields:
- (event_stream_handler)
Returns:
-
(Types::OptimizePromptResponse)
—
Returns a response object which responds to the following methods:
- #optimized_prompt => Types::OptimizedPromptStream
See Also:
2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 2556 def optimize_prompt(params = {}, options = {}, &block) params = params.dup event_stream_handler = case handler = params.delete(:event_stream_handler) when EventStreams::OptimizedPromptStream then handler when Proc then EventStreams::OptimizedPromptStream.new.tap(&handler) when nil then EventStreams::OptimizedPromptStream.new else msg = "expected :event_stream_handler to be a block or "\ "instance of Aws::BedrockAgentRuntime::EventStreams::OptimizedPromptStream"\ ", got `#{handler.inspect}` instead" raise ArgumentError, msg end yield(event_stream_handler) if block_given? req = build_request(:optimize_prompt, params) req.context[:event_stream_handler] = event_stream_handler req.handlers.add(Aws::Binary::DecodeHandler, priority: 95) req.send_request(options, &block) end |
#retrieve(params = {}) ⇒ Types::RetrieveResponse
Queries a knowledge base and retrieves information from it.
The returned response is a pageable response and is Enumerable. For details on usage see PageableResponse.
Examples:
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.retrieve({
knowledge_base_id: "KnowledgeBaseId", # required
next_token: "NextToken",
retrieval_configuration: {
vector_search_configuration: { # required
filter: {
and_all: [
{
# recursive RetrievalFilter
},
],
equals: {
key: "FilterKey", # required
value: { # required
},
},
greater_than: {
key: "FilterKey", # required
value: { # required
},
},
greater_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
in: {
key: "FilterKey", # required
value: { # required
},
},
less_than: {
key: "FilterKey", # required
value: { # required
},
},
less_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
list_contains: {
key: "FilterKey", # required
value: { # required
},
},
not_equals: {
key: "FilterKey", # required
value: { # required
},
},
not_in: {
key: "FilterKey", # required
value: { # required
},
},
or_all: [
{
# recursive RetrievalFilter
},
],
starts_with: {
key: "FilterKey", # required
value: { # required
},
},
string_contains: {
key: "FilterKey", # required
value: { # required
},
},
},
number_of_results: 1,
override_search_type: "HYBRID", # accepts HYBRID, SEMANTIC
},
},
retrieval_query: { # required
text: "KnowledgeBaseQueryTextString", # required
},
})
Response structure
Response structure
resp.next_token #=> String
resp.retrieval_results #=> Array
resp.retrieval_results[0].content.text #=> String
resp.retrieval_results[0].location.confluence_location.url #=> String
resp.retrieval_results[0].location.s3_location.uri #=> String
resp.retrieval_results[0].location.salesforce_location.url #=> String
resp.retrieval_results[0].location.share_point_location.url #=> String
resp.retrieval_results[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
resp.retrieval_results[0].location.web_location.url #=> String
resp.retrieval_results[0].metadata #=> Hash
resp.retrieval_results[0].score #=> Float
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:knowledge_base_id
(required, String)
—
The unique identifier of the knowledge base to query.
-
:next_token
(String)
—
If there are more results than can fit in the response, the response returns a
nextToken
. Use this token in thenextToken
field of another request to retrieve the next batch of results. -
:retrieval_configuration
(Types::KnowledgeBaseRetrievalConfiguration)
—
Contains configurations for the knowledge base query and retrieval process. For more information, see Query configurations.
-
:retrieval_query
(required, Types::KnowledgeBaseQuery)
—
Contains the query to send the knowledge base.
Returns:
-
(Types::RetrieveResponse)
—
Returns a response object which responds to the following methods:
- #next_token => String
- #retrieval_results => Array<Types::KnowledgeBaseRetrievalResult>
See Also:
2708 2709 2710 2711 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 2708 def retrieve(params = {}, options = {}) req = build_request(:retrieve, params) req.send_request(options) end |
#retrieve_and_generate(params = {}) ⇒ Types::RetrieveAndGenerateResponse
Queries a knowledge base and generates responses based on the retrieved results and using the specified foundation model or inference profile. The response only cites sources that are relevant to the query.
Examples:
Request syntax with placeholder values
Request syntax with placeholder values
resp = client.retrieve_and_generate({
input: { # required
text: "RetrieveAndGenerateInputTextString", # required
},
retrieve_and_generate_configuration: {
external_sources_configuration: {
generation_configuration: {
additional_model_request_fields: {
"AdditionalModelRequestFieldsKey" => {
},
},
guardrail_configuration: {
guardrail_id: "GuardrailConfigurationGuardrailIdString", # required
guardrail_version: "GuardrailConfigurationGuardrailVersionString", # required
},
inference_config: {
text_inference_config: {
max_tokens: 1,
stop_sequences: ["RAGStopSequencesMemberString"],
temperature: 1.0,
top_p: 1.0,
},
},
prompt_template: {
text_prompt_template: "TextPromptTemplate",
},
},
model_arn: "BedrockModelArn", # required
sources: [ # required
{
byte_content: {
content_type: "ContentType", # required
data: "data", # required
identifier: "Identifier", # required
},
s3_location: {
uri: "S3Uri", # required
},
source_type: "S3", # required, accepts S3, BYTE_CONTENT
},
],
},
knowledge_base_configuration: {
generation_configuration: {
additional_model_request_fields: {
"AdditionalModelRequestFieldsKey" => {
},
},
guardrail_configuration: {
guardrail_id: "GuardrailConfigurationGuardrailIdString", # required
guardrail_version: "GuardrailConfigurationGuardrailVersionString", # required
},
inference_config: {
text_inference_config: {
max_tokens: 1,
stop_sequences: ["RAGStopSequencesMemberString"],
temperature: 1.0,
top_p: 1.0,
},
},
prompt_template: {
text_prompt_template: "TextPromptTemplate",
},
},
knowledge_base_id: "KnowledgeBaseId", # required
model_arn: "BedrockModelArn", # required
orchestration_configuration: {
additional_model_request_fields: {
"AdditionalModelRequestFieldsKey" => {
},
},
inference_config: {
text_inference_config: {
max_tokens: 1,
stop_sequences: ["RAGStopSequencesMemberString"],
temperature: 1.0,
top_p: 1.0,
},
},
prompt_template: {
text_prompt_template: "TextPromptTemplate",
},
query_transformation_configuration: {
type: "QUERY_DECOMPOSITION", # required, accepts QUERY_DECOMPOSITION
},
},
retrieval_configuration: {
vector_search_configuration: { # required
filter: {
and_all: [
{
# recursive RetrievalFilter
},
],
equals: {
key: "FilterKey", # required
value: { # required
},
},
greater_than: {
key: "FilterKey", # required
value: { # required
},
},
greater_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
in: {
key: "FilterKey", # required
value: { # required
},
},
less_than: {
key: "FilterKey", # required
value: { # required
},
},
less_than_or_equals: {
key: "FilterKey", # required
value: { # required
},
},
list_contains: {
key: "FilterKey", # required
value: { # required
},
},
not_equals: {
key: "FilterKey", # required
value: { # required
},
},
not_in: {
key: "FilterKey", # required
value: { # required
},
},
or_all: [
{
# recursive RetrievalFilter
},
],
starts_with: {
key: "FilterKey", # required
value: { # required
},
},
string_contains: {
key: "FilterKey", # required
value: { # required
},
},
},
number_of_results: 1,
override_search_type: "HYBRID", # accepts HYBRID, SEMANTIC
},
},
},
type: "KNOWLEDGE_BASE", # required, accepts KNOWLEDGE_BASE, EXTERNAL_SOURCES
},
session_configuration: {
kms_key_arn: "KmsKeyArn", # required
},
session_id: "SessionId",
})
Response structure
Response structure
resp.citations #=> Array
resp.citations[0].generated_response_part.text_response_part.span.end #=> Integer
resp.citations[0].generated_response_part.text_response_part.span.start #=> Integer
resp.citations[0].generated_response_part.text_response_part.text #=> String
resp.citations[0].retrieved_references #=> Array
resp.citations[0].retrieved_references[0].content.text #=> String
resp.citations[0].retrieved_references[0].location.confluence_location.url #=> String
resp.citations[0].retrieved_references[0].location.s3_location.uri #=> String
resp.citations[0].retrieved_references[0].location.salesforce_location.url #=> String
resp.citations[0].retrieved_references[0].location.share_point_location.url #=> String
resp.citations[0].retrieved_references[0].location.type #=> String, one of "S3", "WEB", "CONFLUENCE", "SALESFORCE", "SHAREPOINT"
resp.citations[0].retrieved_references[0].location.web_location.url #=> String
resp.citations[0].retrieved_references[0].metadata #=> Hash
resp.guardrail_action #=> String, one of "INTERVENED", "NONE"
resp.output.text #=> String
resp.session_id #=> String
Parameters:
-
params
(Hash)
(defaults to: {})
—
({})
Options Hash (params):
-
:input
(required, Types::RetrieveAndGenerateInput)
—
Contains the query to be made to the knowledge base.
-
:retrieve_and_generate_configuration
(Types::RetrieveAndGenerateConfiguration)
—
Contains configurations for the knowledge base query and retrieval process. For more information, see Query configurations.
-
:session_configuration
(Types::RetrieveAndGenerateSessionConfiguration)
—
Contains details about the session with the knowledge base.
-
:session_id
(String)
—
The unique identifier of the session. When you first make a
RetrieveAndGenerate
request, Amazon Bedrock automatically generates this value. You must reuse this value for all subsequent requests in the same conversational session. This value allows Amazon Bedrock to maintain context and knowledge from previous interactions. You can't explicitly set thesessionId
yourself.
Returns:
-
(Types::RetrieveAndGenerateResponse)
—
Returns a response object which responds to the following methods:
- #citations => Array<Types::Citation>
- #guardrail_action => String
- #output => Types::RetrieveAndGenerateOutput
- #session_id => String
See Also:
2944 2945 2946 2947 |
# File 'gems/aws-sdk-bedrockagentruntime/lib/aws-sdk-bedrockagentruntime/client.rb', line 2944 def retrieve_and_generate(params = {}, options = {}) req = build_request(:retrieve_and_generate, params) req.send_request(options) end |