AWS Bedrock Agents
EcosystemAWS's fully managed AI agent service built on Amazon Bedrock, providing production-grade infrastructure for UCP commerce agents with native AWS security and scaling.
AWS Bedrock Agents is Amazon's managed service for building and deploying AI agents. It runs on Amazon Bedrock, AWS's foundation model platform, and provides managed infrastructure for agent execution, tool calling, session state, and memory. Agents are defined by their instructions and action groups; AWS handles the compute, scaling, and model routing.
For UCP commerce agents, Bedrock Agents' action groups map directly to UCP capabilities. An action group is an OpenAPI schema describing callable endpoints plus a Lambda function that executes the calls. A UCP action group defines the catalog search endpoint, checkout endpoint, and order status endpoint as operations. When the agent decides to call one of these, Bedrock invokes the Lambda function, which executes the HTTP call to the UCP merchant endpoint.
Bedrock Agents integrates with AWS's security infrastructure. IAM roles control which Lambda functions an agent can invoke. AWS Secrets Manager stores UCP payment tokens and identity tokens, preventing them from appearing in agent logs or being accessible outside the Lambda execution context. For enterprise commerce applications where credential management and audit trails are requirements, this native AWS security integration is a significant advantage over self-managed agent deployments.
Bedrock's model selection includes Claude (Anthropic), Llama (Meta), Titan (Amazon), and Mistral, among others. A commerce agent can use Claude Sonnet for complex product comparison tasks and switch to a smaller model for simpler order status lookups, with the model selection managed by Bedrock's routing rules rather than application code.
Why This Matters for UCP
AWS Bedrock Agents provides managed infrastructure and native AWS security (IAM, Secrets Manager) for production UCP commerce agents, handling credential isolation, scaling, and model routing without custom agent server management.
UCP Capabilities
UCP Endpoint
Ecosystem — Endpoint not yet publicly available
Integration Example
// AWS Bedrock Agents: UCP action group + Lambda handler
// Action group OpenAPI schema (stored in S3):
// {
// "openapi": "3.0.0",
// "info": { "title": "UCP Commerce Actions", "version": "1.0" },
// "paths": {
// "/catalog/search": {
// "post": {
// "operationId": "searchCatalog",
// "description": "Search a UCP merchant's product catalog",
// "requestBody": { "content": { "application/json": { "schema": {
// "type": "object",
// "properties": {
// "merchantDomain": { "type": "string" },
// "query": { "type": "string" }
// }
// } } } }
// }
// }
// }
// }
// Lambda handler invoked by Bedrock Agents for each UCP action:
import httpx, json
def handler(event, context):
operation = event["apiPath"]
params = event["requestBody"]["content"]["application/json"]["properties"]
merchant = params["merchantDomain"]["value"]
manifest = httpx.get(f"https://{merchant}/.well-known/ucp").json()
if operation == "/catalog/search":
result = httpx.post(manifest["catalogEndpoint"],
json={"query": params["query"]["value"]}).json()
return {"body": json.dumps(result)}Category
Agent FrameworksUCP Status
EcosystemTags
Pricing
paid
Is this your listing? Claim it to get analytics, a verified badge, and priority placement. Claim your listing →
Reviews
★★★★★4.7 (3 reviews)These are illustrative examples to show what reviews will look like. Submit your own experience below.
R. Kapoor
Platform Engineer · Jan 15, 2026
Integrated in an afternoon. Solid implementation with no unexpected edge cases in production.
J. Santos
Senior Engineer · Feb 28, 2026
Integration was straightforward. The UCP endpoint discovery works as documented and checkout completes in under 2 seconds.
M. Lee
Backend Developer · Mar 3, 2026
Works well for our use case. The documentation is accurate and the implementation matches what the spec describes.
12 other tools in Agent Frameworks
Is this your tool? Claim this listing