Integrate external tools with Amazon Quick Agents using Model Context Protocol (MCP)
Machine Learning Blog
This article provides a comprehensive guide for integrating external tools with Amazon Quick Agents using Model Context Protocol (MCP), enabling ISVs and enterprises to expose application capabilities as MCP tools.
- Amazon Quick MCP integrations support action execution, data access, and AI agent integration
- Six-step checklist: deployment model, MCP server implementation, authentication, documentation, registration, and operations
- Choose between shared multi-tenant or dedicated per-tenant MCP endpoint deployment models
- MCP servers must use HTTPS, support SSE or HTTP streaming, and define tools with JSON schema
- 300-second timeout limit per MCP operation; operations exceeding this fail with HTTP 424
- Support OAuth 2.0, service-to-service, or no authentication based on use case
- Allowlist Amazon Quick OAuth redirect URIs across all AWS regions for identity provider compatibility
- Customers register MCP integration in Amazon Quick console with endpoint URL and authentication details
- Treat MCP servers as production APIs with logging, rate limiting, versioning, and security controls
The guide enables partners to build repeatable integration contracts, define tools once, and support consistent models across customers without custom connectors for each use case.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
2026
2026
2025
2026
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.