D'autres exemples de AWS SDK sont disponibles dans le référentiel [AWS Doc SDK Examples](https://github.com/awsdocs/aws-doc-sdk-examples) GitHub .
Les traductions sont fournies par des outils de traduction automatique. En cas de conflit entre le contenu d'une traduction et celui de la version originale en anglais, la version anglaise prévaudra.
# Exemples de code pour Amazon Bedrock Agents Runtime en utilisant AWS SDKs
Les exemples de code suivants vous montrent comment utiliser Amazon Bedrock Agents Runtime avec un kit de développement AWS logiciel (SDK).
Les *principes de base* sont des exemples de code qui vous montrent comment effectuer les opérations essentielles au sein d’un service.
Les *actions* sont des extraits de code de programmes plus larges et doivent être exécutées dans leur contexte. Alors que les actions vous indiquent comment appeler des fonctions de service individuelles, vous pouvez les voir en contexte dans leurs scénarios associés.
Les *scénarios* sont des exemples de code qui vous montrent comment accomplir des tâches spécifiques en appelant plusieurs fonctions au sein d’un même service ou combinés à d’autres Services AWS.
**Ressources supplémentaires**
+ **[Guide de l’utilisateur de l’exécution d’agents Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/agents.html)** : plus d’informations sur l’exécution d’agents Amazon Bedrock.
+ **[Référence des API de l’exécution d’agents Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_Operations_Agents_for_Amazon_Bedrock_Runtime.html)** : détails sur toutes les actions de l’exécution d’agents Amazon Bedrock disponibles.
+ **[AWS Centre pour les développeurs](https://aws.amazon.com/developer/code-examples/?awsf.sdk-code-examples-product=product%23bedrock-agents)** : exemples de code que vous pouvez filtrer par catégorie ou par recherche en texte intégral.
+ **[AWS Exemples de SDK](https://github.com/awsdocs/aws-doc-sdk-examples)** : GitHub dépôt avec code complet dans les langues préférées. Inclut des instructions sur la configuration et l’exécution du code.
**Contents**
+ [Principes de base](bedrock-agent-runtime_code_examples_basics.md)
+ [Principes de base](bedrock-agent-runtime_example_bedrock-agent-runtime_Scenario_ConverseWithFlow_section.md)
+ [Actions](bedrock-agent-runtime_code_examples_actions.md)
+ [`InvokeAgent`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeAgent_section.md)
+ [`InvokeFlow`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeFlow_section.md)
+ [Scénarios](bedrock-agent-runtime_code_examples_scenarios.md)
+ [Création et invocation d’un flux](bedrock-agent-runtime_example_bedrock-agent_GettingStartedWithBedrockFlows_section.md)
+ [Orchestration des applications d’IA génératives avec Step Functions](bedrock-agent-runtime_example_cross_ServerlessPromptChaining_section.md)
# Exemples de base pour l'utilisation d'Amazon Bedrock Agents Runtime AWS SDKs
Les exemples de code suivants montrent comment utiliser les bases d'Amazon Bedrock Agents Runtime avec AWS SDKs.
**Contents**
+ [Principes de base](bedrock-agent-runtime_example_bedrock-agent-runtime_Scenario_ConverseWithFlow_section.md)
+ [Actions](bedrock-agent-runtime_code_examples_actions.md)
+ [`InvokeAgent`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeAgent_section.md)
+ [`InvokeFlow`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeFlow_section.md)
# Conversation avec un flux Amazon Bedrock
L'exemple de code suivant montre comment InvokeFlow converser avec un flux Amazon Bedrock qui inclut un nœud d'agent.
Pour plus d’informations, consultez [Conversation avec un flux d’Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/flows-multi-turn-invocation.html).
------
#### [ Python ]
**Kit SDK for Python (Boto3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/python/example_code/bedrock-agent-runtime#code-examples).
```
"""
Shows how to run an Amazon Bedrock flow with InvokeFlow and handle muli-turn interaction
for a single conversation.
For more information, see https://docs.aws.amazon.com/bedrock/latest/userguide/flows-multi-turn-invocation.html.
"""
import logging
import boto3
import botocore
import botocore.exceptions
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def invoke_flow(client, flow_id, flow_alias_id, input_data, execution_id):
"""
Invoke an Amazon Bedrock flow and handle the response stream.
Args:
client: Boto3 client for Amazon Bedrock agent runtime.
flow_id: The ID of the flow to invoke.
flow_alias_id: The alias ID of the flow.
input_data: Input data for the flow.
execution_id: Execution ID for continuing a flow. Use the value None on first run.
Returns:
Dict containing flow_complete status, input_required info, and execution_id
"""
response = None
request_params = None
if execution_id is None:
# Don't pass execution ID for first run.
request_params = {
"flowIdentifier": flow_id,
"flowAliasIdentifier": flow_alias_id,
"inputs": [input_data],
"enableTrace": True
}
else:
request_params = {
"flowIdentifier": flow_id,
"flowAliasIdentifier": flow_alias_id,
"executionId": execution_id,
"inputs": [input_data],
"enableTrace": True
}
response = client.invoke_flow(**request_params)
if "executionId" not in request_params:
execution_id = response['executionId']
input_required = None
flow_status = ""
# Process the streaming response
for event in response['responseStream']:
# Check if flow is complete.
if 'flowCompletionEvent' in event:
flow_status = event['flowCompletionEvent']['completionReason']
# Check if more input us needed from user.
elif 'flowMultiTurnInputRequestEvent' in event:
input_required = event
# Print the model output.
elif 'flowOutputEvent' in event:
print(event['flowOutputEvent']['content']['document'])
# Log trace events.
elif 'flowTraceEvent' in event:
logger.info("Flow trace: %s", event['flowTraceEvent'])
return {
"flow_status": flow_status,
"input_required": input_required,
"execution_id": execution_id
}
def converse_with_flow(bedrock_agent_client, flow_id, flow_alias_id):
"""
Run a conversation with the supplied flow.
Args:
bedrock_agent_client: Boto3 client for Amazon Bedrock agent runtime.
flow_id: The ID of the flow to run.
flow_alias_id: The alias ID of the flow.
"""
flow_execution_id = None
finished = False
# Get the intial prompt from the user.
user_input = input("Enter input: ")
# Use prompt to create input data.
flow_input_data = {
"content": {
"document": user_input
},
"nodeName": "FlowInputNode",
"nodeOutputName": "document"
}
try:
while not finished:
# Invoke the flow until successfully finished.
result = invoke_flow(
bedrock_agent_client, flow_id, flow_alias_id, flow_input_data, flow_execution_id)
status = result['flow_status']
flow_execution_id = result['execution_id']
more_input = result['input_required']
if status == "INPUT_REQUIRED":
# The flow needs more information from the user.
logger.info("The flow %s requires more input", flow_id)
user_input = input(
more_input['flowMultiTurnInputRequestEvent']['content']['document'] + ": ")
flow_input_data = {
"content": {
"document": user_input
},
"nodeName": more_input['flowMultiTurnInputRequestEvent']['nodeName'],
"nodeInputName": "agentInputText"
}
elif status == "SUCCESS":
# The flow completed successfully.
finished = True
logger.info("The flow %s successfully completed.", flow_id)
except botocore.exceptions.ClientError as e:
print(f"Client error: {str(e)}")
logger.error("Client error: %s", {str(e)})
except Exception as e:
print(f"An error occurred: {str(e)}")
logger.error("An error occurred: %s", {str(e)})
logger.error("Error type: %s", {type(e)})
def main():
"""
Main entry point for the script.
"""
# Replace these with your actual flow ID and flow alias ID.
FLOW_ID = 'YOUR_FLOW_ID'
FLOW_ALIAS_ID = 'YOUR_FLOW_ALIAS_ID'
logger.info("Starting conversation with FLOW: %s ID: %s",
FLOW_ID, FLOW_ALIAS_ID)
# Get the Bedrock agent runtime client.
session = boto3.Session(profile_name='default')
bedrock_agent_client = session.client('bedrock-agent-runtime')
# Start the conversation.
converse_with_flow(bedrock_agent_client, FLOW_ID, FLOW_ALIAS_ID)
logger.info("Conversation with FLOW: %s ID: %s finished",
FLOW_ID, FLOW_ALIAS_ID)
if __name__ == "__main__":
main()
```
+ Pour plus de détails sur l'API, consultez [InvokeFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-runtime-2023-12-12/InvokeFlow)le *AWS manuel de référence de l'API SDK for Python (Boto3*).
------
# Actions pour Amazon Bedrock Agents Runtime utilisant AWS SDKs
Les exemples de code suivants montrent comment effectuer des actions Amazon Bedrock Agents Runtime individuelles avec AWS SDKs. Chaque exemple inclut un lien vers GitHub, où vous pouvez trouver des instructions pour configurer et exécuter le code.
Ces extraits appellent l’API d’exécution des agents Amazon Bedrock et sont des extraits de code de programmes plus volumineux qui doivent être exécutés en contexte. Vous pouvez voir les actions dans leur contexte dans [Scénarios pour Amazon Bedrock Agents Runtime utilisant AWS SDKs](bedrock-agent-runtime_code_examples_scenarios.md).
Les exemples suivants incluent uniquement les actions les plus couramment utilisées. Pour obtenir la liste complète, consultez la [Référence des API d’exécution des agents Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_Operations_Agents_for_Amazon_Bedrock_Runtime.html).
**Topics**
+ [`InvokeAgent`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeAgent_section.md)
+ [`InvokeFlow`](bedrock-agent-runtime_example_bedrock-agent-runtime_InvokeFlow_section.md)
# Utilisation `InvokeAgent` avec un AWS SDK
Les exemples de code suivants illustrent comment utiliser `InvokeAgent`.
------
#### [ JavaScript ]
**SDK pour JavaScript (v3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/javascriptv3/example_code/bedrock-agent-runtime#code-examples).
```
import {
BedrockAgentRuntimeClient,
InvokeAgentCommand,
} from "@aws-sdk/client-bedrock-agent-runtime";
/**
* @typedef {Object} ResponseBody
* @property {string} completion
*/
/**
* Invokes a Bedrock agent to run an inference using the input
* provided in the request body.
*
* @param {string} prompt - The prompt that you want the Agent to complete.
* @param {string} sessionId - An arbitrary identifier for the session.
*/
export const invokeBedrockAgent = async (prompt, sessionId) => {
const client = new BedrockAgentRuntimeClient({ region: "us-east-1" });
// const client = new BedrockAgentRuntimeClient({
// region: "us-east-1",
// credentials: {
// accessKeyId: "accessKeyId", // permission to invoke agent
// secretAccessKey: "accessKeySecret",
// },
// });
const agentId = "AJBHXXILZN";
const agentAliasId = "AVKP1ITZAA";
const command = new InvokeAgentCommand({
agentId,
agentAliasId,
sessionId,
inputText: prompt,
});
try {
let completion = "";
const response = await client.send(command);
if (response.completion === undefined) {
throw new Error("Completion is undefined");
}
for await (const chunkEvent of response.completion) {
const chunk = chunkEvent.chunk;
console.log(chunk);
const decodedResponse = new TextDecoder("utf-8").decode(chunk.bytes);
completion += decodedResponse;
}
return { sessionId: sessionId, completion };
} catch (err) {
console.error(err);
}
};
// Call function if run directly
import { fileURLToPath } from "node:url";
if (process.argv[1] === fileURLToPath(import.meta.url)) {
const result = await invokeBedrockAgent("I need help.", "123");
console.log(result);
}
```
+ Pour plus de détails sur l'API, reportez-vous [InvokeAgent](https://docs.aws.amazon.com/AWSJavaScriptSDK/v3/latest/client/bedrock-agent-runtime/command/InvokeAgentCommand)à la section *Référence des AWS SDK pour JavaScript API*.
------
#### [ Python ]
**Kit SDK for Python (Boto3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/python/example_code/bedrock-agent-runtime#code-examples).
Invoquez un agent.
```
def invoke_agent(self, agent_id, agent_alias_id, session_id, prompt):
"""
Sends a prompt for the agent to process and respond to.
:param agent_id: The unique identifier of the agent to use.
:param agent_alias_id: The alias of the agent to use.
:param session_id: The unique identifier of the session. Use the same value across requests
to continue the same conversation.
:param prompt: The prompt that you want Claude to complete.
:return: Inference response from the model.
"""
try:
# Note: The execution time depends on the foundation model, complexity of the agent,
# and the length of the prompt. In some cases, it can take up to a minute or more to
# generate a response.
response = self.agents_runtime_client.invoke_agent(
agentId=agent_id,
agentAliasId=agent_alias_id,
sessionId=session_id,
inputText=prompt,
)
completion = ""
for event in response.get("completion"):
chunk = event["chunk"]
completion = completion + chunk["bytes"].decode()
except ClientError as e:
logger.error(f"Couldn't invoke agent. {e}")
raise
return completion
```
+ Pour plus de détails sur l'API, consultez [InvokeAgent](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-runtime-2023-12-12/InvokeAgent)le *AWS manuel de référence de l'API SDK for Python (Boto3*).
------
#### [ Rust ]
**SDK pour Rust**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/rustv1/examples/bedrock-agent-runtime#code-examples).
```
use aws_config::{BehaviorVersion, SdkConfig};
use aws_sdk_bedrockagentruntime::{
self as bedrockagentruntime,
types::{error::ResponseStreamError, ResponseStream},
};
#[allow(unused_imports)]
use mockall::automock;
const BEDROCK_AGENT_ID: &str = "AJBHXXILZN";
const BEDROCK_AGENT_ALIAS_ID: &str = "AVKP1ITZAA";
const BEDROCK_AGENT_REGION: &str = "us-east-1";
#[cfg(not(test))]
pub use EventReceiverImpl as EventReceiver;
#[cfg(test)]
pub use MockEventReceiverImpl as EventReceiver;
pub struct EventReceiverImpl {
inner: aws_sdk_bedrockagentruntime::primitives::event_stream::EventReceiver<
ResponseStream,
ResponseStreamError,
>,
}
#[cfg_attr(test, automock)]
impl EventReceiverImpl {
#[allow(dead_code)]
pub fn new(
inner: aws_sdk_bedrockagentruntime::primitives::event_stream::EventReceiver<
ResponseStream,
ResponseStreamError,
>,
) -> Self {
Self { inner }
}
pub async fn recv(
&mut self,
) -> Result<
Option,
aws_sdk_bedrockagentruntime::error::SdkError<
ResponseStreamError,
aws_smithy_types::event_stream::RawMessage,
>,
> {
self.inner.recv().await
}
}
#[tokio::main]
async fn main() -> Result<(), Box> {
let result = invoke_bedrock_agent("I need help.".to_string(), "123".to_string()).await?;
println!("{}", result);
Ok(())
}
async fn invoke_bedrock_agent(
prompt: String,
session_id: String,
) -> Result {
let sdk_config: SdkConfig = aws_config::defaults(BehaviorVersion::latest())
.region(BEDROCK_AGENT_REGION)
.load()
.await;
let bedrock_client = bedrockagentruntime::Client::new(&sdk_config);
let command_builder = bedrock_client
.invoke_agent()
.agent_id(BEDROCK_AGENT_ID)
.agent_alias_id(BEDROCK_AGENT_ALIAS_ID)
.session_id(session_id)
.input_text(prompt);
let response = command_builder.send().await?;
let response_stream = response.completion;
let event_receiver = EventReceiver::new(response_stream);
process_agent_response_stream(event_receiver).await
}
async fn process_agent_response_stream(
mut event_receiver: EventReceiver,
) -> Result {
let mut full_agent_text_response = String::new();
while let Some(event_result) = event_receiver.recv().await? {
match event_result {
ResponseStream::Chunk(chunk) => {
if let Some(bytes) = chunk.bytes {
match String::from_utf8(bytes.into_inner()) {
Ok(text_chunk) => {
full_agent_text_response.push_str(&text_chunk);
}
Err(e) => {
eprintln!("UTF-8 decoding error for chunk: {}", e);
}
}
}
}
_ => {
panic!("received an unhandled event type from Bedrock stream",);
}
}
}
Ok(full_agent_text_response)
}
#[cfg(test)]
mod test {
use super::*;
#[tokio::test]
async fn test_process_agent_response_stream() {
let mut mock = MockEventReceiverImpl::default();
mock.expect_recv().times(1).returning(|| {
Ok(Some(
aws_sdk_bedrockagentruntime::types::ResponseStream::Chunk(
aws_sdk_bedrockagentruntime::types::PayloadPart::builder()
.set_bytes(Some(aws_smithy_types::Blob::new(vec![
116, 101, 115, 116, 32, 99, 111, 109, 112, 108, 101, 116, 105, 111, 110,
])))
.build(),
),
))
});
// end the stream
mock.expect_recv().times(1).returning(|| Ok(None));
let response = process_agent_response_stream(mock).await.unwrap();
assert_eq!("test completion", response);
}
#[tokio::test]
#[should_panic(expected = "received an unhandled event type from Bedrock stream")]
async fn test_process_agent_response_stream_error() {
let mut mock = MockEventReceiverImpl::default();
mock.expect_recv().times(1).returning(|| {
Ok(Some(
aws_sdk_bedrockagentruntime::types::ResponseStream::Trace(
aws_sdk_bedrockagentruntime::types::TracePart::builder().build(),
),
))
});
let _ = process_agent_response_stream(mock).await.unwrap();
}
}
```
+ Pour plus de détails sur l'API, voir [InvokeAgent](https://docs.rs/aws-sdk-bedrockagentruntime/latest/aws_sdk_bedrockagentruntime/client/struct.Client.html#method.invoke_agent)la section de *référence de l'API AWS SDK for Rust*.
------
#### [ SAP ABAP ]
**Kit SDK pour SAP ABAP**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/sap-abap/services/bdz#code-examples).
```
DATA(lo_result) = lo_bdz->invokeagent(
iv_agentid = iv_agentid
iv_agentaliasid = iv_agentaliasid
iv_enabletrace = abap_true
iv_sessionid = CONV #( cl_system_uuid=>create_uuid_c26_static( ) )
iv_inputtext = |Let's play "rock, paper, scissors". I choose rock.| ).
DATA(lo_stream) = lo_result->get_completion( ).
TRY.
" loop while there are still events in the stream
WHILE lo_stream->/aws1/if_rt_stream_reader~data_available( ) = abap_true.
DATA(lo_evt) = lo_stream->read( ).
" each /AWS1/CL_BDZRESPONSESTREAM_EV event contains exactly one member
" all others are INITIAL. For each event, process the non-initial
" member if desired
IF lo_evt->get_chunk( ) IS NOT INITIAL.
" Process a Chunk event
DATA(lv_xstr) = lo_evt->get_chunk( )->get_bytes( ).
DATA(lv_answer) = /aws1/cl_rt_util=>xstring_to_string( lv_xstr ).
" the answer says something like "I chose paper, so you lost"
ELSEIF lo_evt->get_files( ) IS NOT INITIAL.
" process a Files event if desired
ELSEIF lo_evt->get_returncontrol( ) IS NOT INITIAL.
" process a ReturnControl event if desired
ELSEIF lo_evt->get_trace( ) IS NOT INITIAL.
" process a Trace event if desired
ENDIF.
ENDWHILE.
" the stream of events can possibly contain an exception
" which will be raised to break the loop
" catch /AWS1/CX_BDZACCESSDENIEDEX.
" catch /AWS1/CX_BDZINTERNALSERVEREX.
" catch /AWS1/CX_BDZMODELNOTREADYEX.
" catch /AWS1/CX_BDZVALIDATIONEX.
" catch /AWS1/CX_BDZTHROTTLINGEX.
" catch /AWS1/CX_BDZDEPENDENCYFAILEDEX.
" catch /AWS1/CX_BDZBADGATEWAYEX.
" catch /AWS1/CX_BDZRESOURCENOTFOUNDEX.
" catch /AWS1/CX_BDZSERVICEQUOTAEXCDEX.
" catch /AWS1/CX_BDZCONFLICTEXCEPTION.
ENDTRY.
```
+ Pour plus de détails sur l'API, consultez [InvokeAgent](https://docs.aws.amazon.com/sdk-for-sap-abap/v1/api/latest/index.html)la section de référence du *AWS SDK pour l'API SAP ABAP*.
------
# Utilisation `InvokeFlow` avec un AWS SDK
Les exemples de code suivants illustrent comment utiliser `InvokeFlow`.
Les exemples d’actions sont des extraits de code de programmes de plus grande envergure et doivent être exécutés en contexte. Vous pouvez voir cette action en contexte dans les exemples de code suivants :
+ [Principes de base](bedrock-agent-runtime_example_bedrock-agent-runtime_Scenario_ConverseWithFlow_section.md)
+ [Création et invocation d’un flux](bedrock-agent-runtime_example_bedrock-agent_GettingStartedWithBedrockFlows_section.md)
------
#### [ JavaScript ]
**SDK pour JavaScript (v3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/javascriptv3/example_code/bedrock-agent-runtime#code-examples).
```
import { fileURLToPath } from "node:url";
import {
BedrockAgentRuntimeClient,
InvokeFlowCommand,
} from "@aws-sdk/client-bedrock-agent-runtime";
/**
* Invokes an alias of a flow to run the inputs that you specify and return
* the output of each node as a stream.
*
* @param {{
* flowIdentifier: string,
* flowAliasIdentifier: string,
* prompt?: string,
* region?: string
* }} options
* @returns {Promise} An object containing information about the output from flow invocation.
*/
export const invokeBedrockFlow = async ({
flowIdentifier,
flowAliasIdentifier,
prompt = "Hi, how are you?",
region = "us-east-1",
}) => {
const client = new BedrockAgentRuntimeClient({ region });
const command = new InvokeFlowCommand({
flowIdentifier,
flowAliasIdentifier,
inputs: [
{
content: {
document: prompt,
},
nodeName: "FlowInputNode",
nodeOutputName: "document",
},
],
});
let flowResponse = {};
const response = await client.send(command);
for await (const chunkEvent of response.responseStream) {
const { flowOutputEvent, flowCompletionEvent } = chunkEvent;
if (flowOutputEvent) {
flowResponse = { ...flowResponse, ...flowOutputEvent };
console.log("Flow output event:", flowOutputEvent);
} else if (flowCompletionEvent) {
flowResponse = { ...flowResponse, ...flowCompletionEvent };
console.log("Flow completion event:", flowCompletionEvent);
}
}
return flowResponse;
};
// Call function if run directly
import { parseArgs } from "node:util";
import {
isMain,
validateArgs,
} from "@aws-doc-sdk-examples/lib/utils/util-node.js";
const loadArgs = () => {
const options = {
flowIdentifier: {
type: "string",
required: true,
},
flowAliasIdentifier: {
type: "string",
required: true,
},
prompt: {
type: "string",
},
region: {
type: "string",
},
};
const results = parseArgs({ options });
const { errors } = validateArgs({ options }, results);
return { errors, results };
};
if (isMain(import.meta.url)) {
const { errors, results } = loadArgs();
if (!errors) {
invokeBedrockFlow(results.values);
} else {
console.error(errors.join("\n"));
}
}
```
+ Pour plus de détails sur l'API, reportez-vous [InvokeFlow](https://docs.aws.amazon.com/AWSJavaScriptSDK/v3/latest/client/bedrock-agent-runtime/command/InvokeFlowCommand)à la section *Référence des AWS SDK pour JavaScript API*.
------
#### [ Python ]
**Kit SDK for Python (Boto3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/python/example_code/bedrock-agent-runtime#code-examples).
Invoquez un flux.
```
def invoke_flow(self, flow_id, flow_alias_id, input_data, execution_id):
"""
Invoke an Amazon Bedrock flow and handle the response stream.
Args:
param flow_id: The ID of the flow to invoke.
param flow_alias_id: The alias ID of the flow.
param input_data: Input data for the flow.
param execution_id: Execution ID for continuing a flow. Use the value None on first run.
Return: Response from the flow.
"""
try:
request_params = None
if execution_id is None:
# Don't pass execution ID for first run.
request_params = {
"flowIdentifier": flow_id,
"flowAliasIdentifier": flow_alias_id,
"inputs": input_data,
"enableTrace": True
}
else:
request_params = {
"flowIdentifier": flow_id,
"flowAliasIdentifier": flow_alias_id,
"executionId": execution_id,
"inputs": input_data,
"enableTrace": True
}
response = self.agents_runtime_client.invoke_flow(**request_params)
if "executionId" not in request_params:
execution_id = response['executionId']
result = ""
# Get the streaming response
for event in response['responseStream']:
result = result + str(event) + '\n'
print(result)
except ClientError as e:
logger.error("Couldn't invoke flow %s.", {e})
raise
return result
```
+ Pour plus de détails sur l'API, consultez [InvokeFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-runtime-2023-12-12/InvokeFlow)le *AWS manuel de référence de l'API SDK for Python (Boto3*).
------
# Scénarios pour Amazon Bedrock Agents Runtime utilisant AWS SDKs
Les exemples de code suivants vous montrent comment implémenter des scénarios courants dans Amazon Bedrock Agents Runtime avec AWS SDKs. Ces scénarios vous montrent comment accomplir des tâches spécifiques en appelant plusieurs fonctions d’exécution des agents Amazon Bedrock ou en les combinant avec d’autres Services AWS. Chaque exemple inclut un lien vers le code source complet, où vous trouverez des instructions sur la configuration et l’exécution du code.
Les scénarios ciblent un niveau d’expérience intermédiaire pour vous aider à comprendre les actions de service dans leur contexte.
**Topics**
+ [Création et invocation d’un flux](bedrock-agent-runtime_example_bedrock-agent_GettingStartedWithBedrockFlows_section.md)
+ [Orchestration des applications d’IA génératives avec Step Functions](bedrock-agent-runtime_example_cross_ServerlessPromptChaining_section.md)
# end-to-endExemple montrant comment créer et invoquer un flux Amazon Bedrock à l'aide d'un SDK AWS
L’exemple de code suivant illustre comment :
+ créer un rôle d’exécution pour le flux ;
+ créer le flux ;
+ déployer le flux entièrement configuré ;
+ invoquer le flux à l’aide des invites fournies par l’utilisateur ;
+ Supprimez toutes les ressources créées.
------
#### [ Python ]
**Kit SDK for Python (Boto3)**
Il y en a plus à ce sujet GitHub. Trouvez l’exemple complet et découvrez comment le configurer et l’exécuter dans le [référentiel d’exemples de code AWS](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/python/example_code/bedrock-agent#code-examples).
Génère une liste de lecture musicale en fonction du genre et du nombre de chansons spécifiés par l’utilisateur.
```
from datetime import datetime
import logging
import boto3
from botocore.exceptions import ClientError
from roles import create_flow_role, delete_flow_role, update_role_policy
from flow import create_flow, prepare_flow, delete_flow
from run_flow import run_playlist_flow
from flow_version import create_flow_version, delete_flow_version
from flow_alias import create_flow_alias, delete_flow_alias
logging.basicConfig(
level=logging.INFO
)
logger = logging.getLogger(__name__)
def create_input_node(name):
"""
Creates an input node configuration for an Amazon Bedrock flow.
The input node serves as the entry point for the flow and defines
the initial document structure that will be passed to subsequent nodes.
Args:
name (str): The name of the input node.
Returns:
dict: The input node configuration.
"""
return {
"type": "Input",
"name": name,
"outputs": [
{
"name": "document",
"type": "Object"
}
]
}
def create_prompt_node(name, model_id):
"""
Creates a prompt node configuration for a Bedrock flow that generates music playlists.
The prompt node defines an inline prompt template that creates a music playlist based on
a specified genre and number of songs. The prompt uses two variables that are mapped from
the input JSON object:
- {{genre}}: The genre of music to create a playlist for
- {{number}}: The number of songs to include in the playlist
Args:
name (str): The name of the prompt node.
model_id (str): The identifier of the foundation model to use for the prompt.
Returns:
dict: The prompt node.
"""
return {
"type": "Prompt",
"name": name,
"configuration": {
"prompt": {
"sourceConfiguration": {
"inline": {
"modelId": model_id,
"templateType": "TEXT",
"inferenceConfiguration": {
"text": {
"temperature": 0.8
}
},
"templateConfiguration": {
"text": {
"text": "Make me a {{genre}} playlist consisting of the following number of songs: {{number}}."
}
}
}
}
}
},
"inputs": [
{
"name": "genre",
"type": "String",
"expression": "$.data.genre"
},
{
"name": "number",
"type": "Number",
"expression": "$.data.number"
}
],
"outputs": [
{
"name": "modelCompletion",
"type": "String"
}
]
}
def create_output_node(name):
"""
Creates an output node configuration for a Bedrock flow.
The output node validates that the output from the last node is a string
and returns it unmodified. The input name must be "document".
Args:
name (str): The name of the output node.
Returns:
dict: The output node configuration containing the output node:
"""
return {
"type": "Output",
"name": name,
"inputs": [
{
"name": "document",
"type": "String",
"expression": "$.data"
}
]
}
def create_playlist_flow(client, flow_name, flow_description, role_arn, prompt_model_id):
"""
Creates the playlist generator flow.
Args:
client: bedrock agent boto3 client.
role_arn (str): Name for the new IAM role.
prompt_model_id (str): The id of the model to use in the prompt node.
Returns:
dict: The response from the create_flow operation.
"""
input_node = create_input_node("FlowInput")
prompt_node = create_prompt_node("MakePlaylist", prompt_model_id)
output_node = create_output_node("FlowOutput")
# Create connections between the nodes
connections = []
# First, create connections between the output of the flow
# input node and each input of the prompt node.
for prompt_node_input in prompt_node["inputs"]:
connections.append(
{
"name": "_".join([input_node["name"], prompt_node["name"],
prompt_node_input["name"]]),
"source": input_node["name"],
"target": prompt_node["name"],
"type": "Data",
"configuration": {
"data": {
"sourceOutput": input_node["outputs"][0]["name"],
"targetInput": prompt_node_input["name"]
}
}
}
)
# Then, create a connection between the output of the prompt node and the input of the flow output node
connections.append(
{
"name": "_".join([prompt_node["name"], output_node["name"]]),
"source": prompt_node["name"],
"target": output_node["name"],
"type": "Data",
"configuration": {
"data": {
"sourceOutput": prompt_node["outputs"][0]["name"],
"targetInput": output_node["inputs"][0]["name"]
}
}
}
)
flow_def = {
"nodes": [input_node, prompt_node, output_node],
"connections": connections
}
# Create the flow.
response = create_flow(
client, flow_name, flow_description, role_arn, flow_def)
return response
def get_model_arn(client, model_id):
"""
Gets the Amazon Resource Name (ARN) for a model.
Args:
client (str): Amazon Bedrock boto3 client.
model_id (str): The id of the model.
Returns:
str: The ARN of the model.
"""
try:
# Call GetFoundationModelDetails operation
response = client.get_foundation_model(modelIdentifier=model_id)
# Extract model ARN from the response
model_arn = response['modelDetails']['modelArn']
return model_arn
except ClientError as e:
logger.exception("Client error getting model ARN: %s", {str(e)})
raise
except Exception as e:
logger.exception("Unexpected error getting model ARN: %s", {str(e)})
raise
def prepare_flow_version_and_alias(bedrock_agent_client,
flow_id):
"""
Prepares the flow and then creates a flow version and flow alias.
Args:
bedrock_agent_client: Amazon Bedrock Agent boto3 client.
flowd_id (str): The ID of the flow that you want to prepare.
Returns: The flow_version and flow_alias.
"""
status = prepare_flow(bedrock_agent_client, flow_id)
flow_version = None
flow_alias = None
if status == 'Prepared':
# Create the flow version and alias.
flow_version = create_flow_version(bedrock_agent_client,
flow_id,
f"flow version for flow {flow_id}.")
flow_alias = create_flow_alias(bedrock_agent_client,
flow_id,
flow_version,
"latest",
f"Alias for flow {flow_id}, version {flow_version}")
return flow_version, flow_alias
def delete_role_resources(bedrock_agent_client,
iam_client,
role_name,
flow_id,
flow_version,
flow_alias):
"""
Deletes the flow, flow alias, flow version, and IAM roles.
Args:
bedrock_agent_client: Amazon Bedrock Agent boto3 client.
iam_client: Amazon IAM boto3 client.
role_name (str): The name of the IAM role.
flow_id (str): The id of the flow.
flow_version (str): The version of the flow.
flow_alias (str): The alias of the flow.
"""
if flow_id is not None:
if flow_alias is not None:
delete_flow_alias(bedrock_agent_client, flow_id, flow_alias)
if flow_version is not None:
delete_flow_version(bedrock_agent_client,
flow_id, flow_version)
delete_flow(bedrock_agent_client, flow_id)
if role_name is not None:
delete_flow_role(iam_client, role_name)
def main():
"""
Creates, runs, and optionally deletes a Bedrock flow for generating music playlists.
Note:
Requires valid AWS credentials in the default profile
"""
delete_choice = "y"
try:
# Get various boto3 clients.
session = boto3.Session(profile_name='default')
bedrock_agent_runtime_client = session.client('bedrock-agent-runtime')
bedrock_agent_client = session.client('bedrock-agent')
bedrock_client = session.client('bedrock')
iam_client = session.client('iam')
role_name = None
flow_id = None
flow_version = None
flow_alias = None
#Change the model as needed.
prompt_model_id = "amazon.nova-pro-v1:0"
# Base the flow name on the current date and time
current_time = datetime.now()
timestamp = current_time.strftime("%Y-%m-%d-%H-%M-%S")
flow_name = f"FlowPlayList_{timestamp}"
flow_description = "A flow to generate a music playlist."
# Create a role for the flow.
role_name = f"BedrockFlowRole-{flow_name}"
role = create_flow_role(iam_client, role_name)
role_arn = role['Arn']
# Create the flow.
response = create_playlist_flow(
bedrock_agent_client, flow_name, flow_description, role_arn, prompt_model_id)
flow_id = response.get('id')
if flow_id:
# Update accessible resources in the role.
model_arn = get_model_arn(bedrock_client, prompt_model_id)
update_role_policy(iam_client, role_name, [
response.get('arn'), model_arn])
# Prepare the flow and flow version.
flow_version, flow_alias = prepare_flow_version_and_alias(
bedrock_agent_client, flow_id)
# Run the flow.
if flow_version and flow_alias:
run_playlist_flow(bedrock_agent_runtime_client,
flow_id, flow_alias)
delete_choice = input("Delete flow? y or n : ").lower()
else:
print("Couldn't run. Deleting flow and role.")
delete_flow(bedrock_agent_client, flow_id)
delete_flow_role(iam_client, role_name)
else:
print("Couldn't create flow.")
except Exception as e:
print(f"Fatal error: {str(e)}")
finally:
if delete_choice == 'y':
delete_role_resources(bedrock_agent_client,
iam_client,
role_name,
flow_id,
flow_version,
flow_alias)
else:
print("Flow not deleted. ")
print(f"\tFlow ID: {flow_id}")
print(f"\tFlow version: {flow_version}")
print(f"\tFlow alias: {flow_alias}")
print(f"\tRole ARN: {role_arn}")
print("Done!")
if __name__ == "__main__":
main()
def invoke_flow(client, flow_id, flow_alias_id, input_data):
"""
Invoke an Amazon Bedrock flow and handle the response stream.
Args:
client: Boto3 client for Amazon Bedrock agent runtime.
flow_id: The ID of the flow to invoke.
flow_alias_id: The alias ID of the flow.
input_data: Input data for the flow.
Returns:
Dict containing flow status and flow output.
"""
response = None
request_params = None
request_params = {
"flowIdentifier": flow_id,
"flowAliasIdentifier": flow_alias_id,
"inputs": [input_data],
"enableTrace": True
}
response = client.invoke_flow(**request_params)
flow_status = ""
output= ""
# Process the streaming response
for event in response['responseStream']:
# Check if flow is complete.
if 'flowCompletionEvent' in event:
flow_status = event['flowCompletionEvent']['completionReason']
# Save the model output.
elif 'flowOutputEvent' in event:
output = event['flowOutputEvent']['content']['document']
logger.info("Output : %s", output)
# Log trace events.
elif 'flowTraceEvent' in event:
logger.info("Flow trace: %s", event['flowTraceEvent'])
return {
"flow_status": flow_status,
"output": output
}
def run_playlist_flow(bedrock_agent_client, flow_id, flow_alias_id):
"""
Runs the playlist generator flow.
Args:
bedrock_agent_client: Boto3 client for Amazon Bedrock agent runtime.
flow_id: The ID of the flow to run.
flow_alias_id: The alias ID of the flow.
"""
print ("Welcome to the playlist generator flow.")
# Get the initial prompt from the user.
genre = input("Enter genre: ")
number_of_songs = int(input("Enter number of songs: "))
# Use prompt to create input data for the input node.
flow_input_data = {
"content": {
"document": {
"genre" : genre,
"number" : number_of_songs
}
},
"nodeName": "FlowInput",
"nodeOutputName": "document"
}
try:
result = invoke_flow(
bedrock_agent_client, flow_id, flow_alias_id, flow_input_data)
status = result['flow_status']
if status == "SUCCESS":
# The flow completed successfully.
logger.info("The flow %s successfully completed.", flow_id)
print(result['output'])
else:
logger.warning("Flow status: %s",status)
except ClientError as e:
print(f"Client error: {str(e)}")
logger.error("Client error: %s", {str(e)})
raise
except Exception as e:
logger.error("An error occurred: %s", {str(e)})
logger.error("Error type: %s", {type(e)})
raise
def create_flow_role(client, role_name):
"""
Creates an IAM role for Amazon Bedrock with permissions to run a flow.
Args:
role_name (str): Name for the new IAM role.
Returns:
str: The role Amazon Resource Name.
"""
# Trust relationship policy - allows Amazon Bedrock service to assume this role.
trust_policy = {
"Version":"2012-10-17",
"Statement": [{
"Effect": "Allow",
"Principal": {
"Service": "bedrock.amazonaws.com"
},
"Action": "sts:AssumeRole"
}]
}
# Basic inline policy for for running a flow.
resources = "*"
bedrock_policy = {
"Version":"2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"bedrock:InvokeModel",
"bedrock:Retrieve",
"bedrock:RetrieveAndGenerate"
],
# Using * as placeholder - Later you update with specific ARNs.
"Resource": resources
}
]
}
try:
# Create the IAM role with trust policy
logging.info("Creating role: %s",role_name)
role = client.create_role(
RoleName=role_name,
AssumeRolePolicyDocument=json.dumps(trust_policy),
Description="Role for Amazon Bedrock operations"
)
# Attach inline policy to the role
print("Attaching inline policy")
client.put_role_policy(
RoleName=role_name,
PolicyName=f"{role_name}-policy",
PolicyDocument=json.dumps(bedrock_policy)
)
logging.info("Create Role ARN: %s", role['Role']['Arn'])
return role['Role']
except ClientError as e:
logging.warning("Error creating role: %s", str(e))
raise
except Exception as e:
logging.warning("Unexpected error: %s", str(e))
raise
def update_role_policy(client, role_name, resource_arns):
"""
Updates an IAM role's inline policy with specific resource ARNs.
Args:
role_name (str): Name of the existing role.
resource_arns (list): List of resource ARNs to allow access to.
"""
updated_policy = {
"Version":"2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"bedrock:GetFlow",
"bedrock:InvokeModel",
"bedrock:Retrieve",
"bedrock:RetrieveAndGenerate"
],
"Resource": resource_arns
}
]
}
try:
client.put_role_policy(
RoleName=role_name,
PolicyName=f"{role_name}-policy",
PolicyDocument=json.dumps(updated_policy)
)
logging.info("Updated policy for role: %s",role_name)
except ClientError as e:
logging.warning("Error updating role policy: %s", str(e))
raise
def delete_flow_role(client, role_name):
"""
Deletes an IAM role.
Args:
role_name (str): Name of the role to delete.
"""
try:
# Detach and delete inline policies
policies = client.list_role_policies(RoleName=role_name)['PolicyNames']
for policy_name in policies:
client.delete_role_policy(RoleName=role_name, PolicyName=policy_name)
# Delete the role
client.delete_role(RoleName=role_name)
logging.info("Deleted role: %s", role_name)
except ClientError as e:
logging.info("Error Deleting role: %s", str(e))
raise
```
+ Pour plus de détails sur l’API, consultez les rubriques suivantes dans la *Référence des API du kit AWS SDK for Python (Boto3)*.
+ [CreateFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/CreateFlow)
+ [CreateFlowAlias](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/CreateFlowAlias)
+ [CreateFlowVersion](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/CreateFlowVersion)
+ [DeleteFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/DeleteFlow)
+ [DeleteFlowAlias](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/DeleteFlowAlias)
+ [DeleteFlowVersion](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/DeleteFlowVersion)
+ [GetFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/GetFlow)
+ [GetFlowAlias](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/GetFlowAlias)
+ [GetFlowVersion](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/GetFlowVersion)
+ [InvokeFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-runtime-2023-12-12/InvokeFlow)
+ [PrepareFlow](https://docs.aws.amazon.com/goto/boto3/bedrock-agent-2023-12-12/PrepareFlow)
------
# Création et orchestration d’applications d’IA générative avec Amazon Bedrock et Step Functions
L’exemple de code suivant montre comment créer et orchestrer des applications d’IA générative avec Amazon Bedrock et Step Functions.
------
#### [ Python ]
**Kit SDK for Python (Boto3)**
Le scénario d’enchaînement des invites Amazon Bedrock sans serveur montre comment [AWS Step Functions](https://docs.aws.amazon.com/step-functions/latest/dg/welcome.html), [Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/what-is-bedrock.html) et [https://docs.aws.amazon.com/bedrock/latest/userguide/agents.html](https://docs.aws.amazon.com/bedrock/latest/userguide/agents.html) peuvent être utilisés pour créer et orchestrer des applications d’IA générative complexes, sans serveur et hautement évolutives. Il contient les exemples pratiques suivants :
+ Rédigez une analyse d’un roman donné pour un blog littéraire. Cet exemple illustre une chaîne d’invites simple et séquentielle.
+ Générez une courte histoire sur un sujet donné. Cet exemple illustre comment l’IA peut traiter de manière itérative une liste d’éléments qu’elle a précédemment générée.
+ Créez un itinéraire pour un week-end de vacances vers une destination donnée. Cet exemple illustre comment paralléliser plusieurs invites distinctes.
+ Présentez des idées de films à un utilisateur humain agissant en tant que producteur de films. Cet exemple illustre comment paralléliser la même invite avec différents paramètres d’inférence, comment revenir à une étape précédente de la chaîne et comment inclure une entrée humaine dans le flux de travail.
+ Planifiez un repas en fonction des ingrédients que l’utilisateur a à portée de main. Cet exemple illustre comment les enchaînements des invites peuvent intégrer deux conversations distinctes basées sur l’IA, avec deux personnages d’IA engageant un débat entre eux pour améliorer le résultat final.
+ Trouvez et résumez le GitHub référentiel le plus populaire du moment. Cet exemple illustre le chaînage de plusieurs agents d'IA qui interagissent avec des agents externes APIs.
Pour le code source complet et les instructions de configuration et d'exécution, consultez le projet complet sur [GitHub](https://github.com/aws-samples/amazon-bedrock-serverless-prompt-chaining).
**Les services utilisés dans cet exemple**
+ Amazon Bedrock
+ Exécution d’Amazon Bedrock
+ Agents Amazon Bedrock
+ Exécution des agents Amazon Bedrock
+ Step Functions
------