Artificial Intelligence

Smartsheet Launches MCP Server to Connect AI Agents on AWS

The newly released Smartsheet Model Context Protocol Server allows AI assistants to read, write, and manage live work management data across platforms without requiring users to switch tools.

Radwan AbdulhadiEditor and technology writer

Published: October 20, 2026 · Updated: October 20, 2026 · 3 min read

Smartsheet Launches MCP Server to Connect AI Agents on AWS — Image: AWS Machine Learning Blog

Introduction to Smartsheet MCP Server

AI agents are most useful when they can act on real work happening across an organization. Work management typically involves complex, cross-functional tasks like updating project timelines, reassigning tasks, and surfacing status. To support these workflows without requiring users to switch tools, <a href="https://www.smartsheet.com/">Smartsheet</a> has launched a generally available Model Context Protocol (MCP) Server.

With <a href="https://developers.smartsheet.com/ai-mcp/smartsheet/mcp-server">the Smartsheet MCP Server</a>, AI assistants can read, write, and manage work data through a standardized protocol. The protocol covers more than 40 tools designed for exploring, creating, and managing sheets, rows, columns, and automations directly through natural language.

Architecture diagram: the Smartsheet MCP Server connecting Amazon Quick and Kiro clients to the Smartsheet API and data layer — Image related to the report from AWS Machine Learning Blog · Source: AWS Machine Learning Blog

Integration Paths on AWS: Amazon Quick and Kiro

The platform introduces two primary ways to connect the MCP server within existing AWS workflows. <a href="https://aws.amazon.com/quick/">Amazon Quick</a> is an agentic AI workspace built for business users, project managers, and operations teams who want to ask questions, update sheets, and build dashboards through natural language without coding.

Alternatively, <a href="https://kiro.dev/">Kiro</a> serves as an agentic IDE where developers connect to project management tools via MCP. This allows them to reference project specs, update task statuses, and track dependencies without leaving their coding workflow.

Amazon Quick Connectors directory with the Smartsheet connector tile — Image related to the report from AWS Machine Learning Blog · Source: AWS Machine Learning Blog

Workflow Scenarios for Teams

Consider a software engineering team tracking sprints using project sheets. Project managers can use Amazon Quick through natural language to check which sprint tasks are blocked during a given week. The chat agent queries the central MCP server, pulls relevant rows, and displays a summary containing owner names and due dates derived directly from live project data.

Meanwhile, developers fixing bugs can ask Kiro to pull the current sprint sheet and display open backend tasks assigned to them. Kiro queries the project management data inline, and once a fix is committed, a Kiro hook automatically updates the corresponding row status to complete.

Smartsheet connection details form in Amazon Quick with base URL, client ID, client secret, token URL, and auth URL fields — Image related to the report from AWS Machine Learning Blog · Source: AWS Machine Learning Blog

Engineering and Architecture Details

The integration layer delivers structured context that includes row relationships, column types, cell history, and comments. This design produces token-efficient payload sizes rather than returning raw API payloads, helping enterprise users reduce costs and improve latency at scale when handling hundreds of AI-assisted queries across an entire portfolio.

For technical implementation details regarding infrastructure, developers can review how the remote architecture was established by reading <a href="https://aws.amazon.com/blogs/machine-learning/how-smartsheet-built-a-remote-mcp-server-on-aws/">How Smartsheet built a remote MCP server on AWS</a>.

Manage Tools and Permissions page listing the available Smartsheet tools in Amazon Quick — Image related to the report from AWS Machine Learning Blog · Source: AWS Machine Learning Blog

Prerequisites and Configuration

Before connecting the server, organizations need an active AWS account with appropriate IAM permissions, a Business, Enterprise, or AWM Smartsheet plan, and a Personal Access Token or OAuth credentials. Business users require appropriate permissions inside the <a href="https://docs.aws.amazon.com/quicksuite/latest/userguide/mcp-integration.html">Amazon Quick console</a>, while developers need the IDE installed from <a href="https://kiro.dev">kiro.dev</a> along with Node.js.

The server operates across three regional endpoints using Streamable HTTP transport and authenticates via OAuth 2.0 or Personal Access Tokens. Setup requires configuring connection attributes such as the base endpoint at <a href="https://mcp.smartsheet.com">https://mcp.smartsheet.com</a> and token authentication routing through <a href="https://api.smartsheet.com/2.0/token">https://api.smartsheet.com/2.0/token</a>.