An AWS guide walks through connecting MCP servers hosted on AgentCore Runtime to Amazon Quick, letting foundation models access external data and tools through standardized, secure integrations. The post frames MCP as the emerging standard for agentic AI workflows, enabling stateful, multi-turn agent capabilities with reduced hallucination via accurate context.
An AWS tutorial/pattern for building AI agents that reason over data spread across multiple AWS accounts using AgentCore Gateway and the Model Context Protocol. The LLM agent queries data in each line-of-business account in place, avoiding data replication or complex cross-account IAM setups. It targets enterprises wanting agents with visibility across distributed accounts.
An AWS tutorial/pattern for building AI agents that reason over data spread across multiple AWS accounts using AgentCore Gateway and the Model Context Protocol. The LLM agent queries data in each line-of-business account in place, avoiding data replication or complex cross-account IAM setups. It targets enterprises wanting agents with visibility across distributed accounts.
An AWS blog post details a defense-in-depth authorization pattern for Model Context Protocol (MCP) tool invocations on Amazon Quick. A single AWS Lambda REQUEST interceptor attached to an Amazon Bedrock AgentCore Gateway evaluates OIDC JWT claims through four sequential gates — MFA verification, country geo-fencing, group-based RBAC, and tool permission checks — before business logic runs. The walkthrough uses Microsoft Entra ID as the identity provider and targets regulated organizations needing granular access control for compliance.
An AWS blog post details a defense-in-depth authorization pattern for Model Context Protocol (MCP) tool invocations on Amazon Quick. A single AWS Lambda REQUEST interceptor attached to an Amazon Bedrock AgentCore Gateway evaluates OIDC JWT claims through four sequential gates — MFA verification, country geo-fencing, group-based RBAC, and tool permission checks — before business logic runs. The walkthrough uses Microsoft Entra ID as the identity provider and targets regulated organizations needing granular access control for compliance.
AgentCore engineers published a walkthrough of an MCP bridge that lets their cloud-hosted AI agent access local MCP tools, such as spreadsheets on users' laptops. The bridge leverages the Model Context Protocol, an open standard introduced by Anthropic in November 2024, to standardize AI connections to external data and tools.
AgentCore engineers published a walkthrough of an MCP bridge that lets their cloud-hosted AI agent access local MCP tools, such as spreadsheets on users' laptops. The bridge leverages the Model Context Protocol, an open standard introduced by Anthropic in November 2024, to standardize AI connections to external data and tools.
A reference architecture for public sector agencies that turns unstructured evidence like body camera footage and scanned documents into structured insights. Amazon Bedrock Data Automation extracts the insights, an MCP layer exposes them, and Salesforce Agentforce queries them via natural language. It solves the problem of acting on large volumes of unstructured evidence without manual review.
A reference architecture for public sector agencies that turns unstructured evidence like body camera footage and scanned documents into structured insights. Amazon Bedrock Data Automation extracts the insights, an MCP layer exposes them, and Salesforce Agentforce queries them via natural language. It solves the problem of acting on large volumes of unstructured evidence without manual review.
Smartsheet built a remote Model Context Protocol (MCP) server on AWS, enabling AI assistants like Amazon Quick and Claude Desktop to directly access Smartsheet's enterprise work management data. This bridges the gap between AI agents and structured data in enterprise systems.
Smartsheet built a remote Model Context Protocol (MCP) server on AWS, enabling AI assistants like Amazon Quick and Claude Desktop to directly access Smartsheet's enterprise work management data. This bridges the gap between AI agents and structured data in enterprise systems.