Oriel — AI & technology worth noticing, every day.

How MoEngage turned over 400 APIs into agent tools with AgentCore Gateway

Web page · https://aws.amazon.com/blogs/machine-learning/how-moengage-turned-over-400-apis-into-agent-tools-with-agentcore-gateway

MoEngage turned roughly 400 private REST APIs across 7 AWS Regions into permission-scoped agent tools with Amazon Bedrock AgentCore Gateway. No per-service MCP servers, no public endpoints, and no 400-tool context bloat. See how OpenAPI specs, Cedar policies, and managed Amazon VPC Lattice egress made it work.

First seen by Oriel on .

AI summary

MoEngage used Amazon Bedrock AgentCore Gateway to expose roughly 400 private REST API operations as permission-scoped MCP tools across seven AWS Regions without building MCP servers or public endpoints.

  • Teams annotate existing OpenAPI specifications with x-moe-mcp fields; AgentCore Gateway generates and routes MCP tools, while CI handles registration and Cedar policy application.
  • Cedar policies scope tools server-side by agent IAM role, so unauthorized write operations are omitted from tools/list rather than filtered by clients.
  • AgentCore managed egress over Amazon VPC Lattice connects privately to internal APIs through VPC interface endpoints, avoiding per-API networking changes.
  • Semantic search and a client-side search_tools meta-tool keep hundreds of definitions out of model context and let agents discover capabilities by intent.
  • The production deployment serves 10 internal agents, saved approximately six months of engineering effort, and reduced new API onboarding from days to hours.

#amazon bedrock agentcore #mcp #cedar #api gateways

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
AWS Machine Learning Blog1638ab30d37c69083638027cdd108a42db1e855ahttps://aws.amazon.com/blogs/machine-learning/how-moengage-turned-over-400-apis-into-agent-tools-with-agentcore-gateway/ Read discussion