As AWS infrastructure scales, operational workflows naturally grow more complex. SREs and DevOps engineers spend significant time context-switching between the AWS Management Console, CLI documentation, and multiple service dashboards. They manually translate business questions into the correct API syntax, chain calls across services, and rebuild the same integration patterns for each new use case. This friction compounds during incident investigations, requiring cross-referencing of Amazon CloudWatch logs, Amazon EC2 instance states, and IAM policies across separate interfaces.
This post introduces how to use Amazon Bedrock AgentCore Runtime with Model Context Protocol (MCP) support to connect Amazon Quick—a conversational AI interface—with AWS services through the AWS API MCP Server. This architecture creates an AI assistant that translates natural language into AWS Command Line Interface (AWS CLI) commands, eliminating the need to switch between tools during critical moments.
With Bedrock AgentCore Runtime and MCP support, natural language queries translate directly to AWS API calls (e.g., "Show running EC2 instances in us-east-1") securely within existing IAM permissions. Rather than rebuilding connection logic for each workflow, teams can standardize how AI agents interact with AWS services through a single, reusable integration.
The operational workflow runs as follows:
1. User Query: The user asks a natural language question to the Amazon Quick custom agent.
2. Cognito Authentication: Amazon Quick obtains a JWT token from your Amazon Cognito user pool using OAuth 2.0 client credentials flow.
3. Token Validation: The request reaches Amazon Bedrock AgentCore Runtime, which validates the JWT token against the Cognito configuration.
4. Routing & Translation: AgentCore Runtime securely invokes the AWS API MCP Server running in a containerized environment, which translates the natural language query into the appropriate AWS CLI command.
5. Execution: Using the preconfigured IAM execution role, AWS services execute the command and return the results.
[AgentUpdate Depth Analysis] AWS's integration of the Anthropic-pioneered Model Context Protocol (MCP) into Bedrock AgentCore represents a significant milestone in the AI Agent ecosystem. Historically, integrating LLMs with complex enterprise cloud environments required highly fragmented custom Tool or Plugin architectures, posing maintenance challenges and limiting agent portability. By adopting MCP, AWS establishes a standardized interface for agents to discover and interact with cloud resources. This shift from bespoke API wrappers to unified protocol-based orchestration will accelerate the adoption of AIOps, turning operations into a declarative natural-language interface and establishing MCP as the industry standard for LLM-to-infrastructure integration.
