Skip to content
AI Radar by Wagner Solution
CS
Menu
worth noting AI agents

MoEngage made approximately 400 internal REST APIs available to agents through AgentCore Gateway

only one source so far

MoEngage used AgentCore Gateway to convert approximately 400 internal REST APIs into tools for AI agents. Product teams provide OpenAPI specifications, while a central team manages the MCP layer. The company estimates savings of six months of platform development.

MoEngage used AgentCore Gateway, part of Amazon Bedrock AgentCore, to make approximately 400 internal REST APIs available as tools for AI agents. These APIs are managed by 25 development teams across seven AWS regions. According to the company, connectivity remains private and does not require publicly accessible endpoints.

Product teams provide OpenAPI specifications and determine which operations should be available as tools. AgentCore Gateway creates MCP tools from these specifications and routes calls to the appropriate services. A central team manages the MCP layer, including permissions for individual agents; product teams do not write their own server-side code for MCP.

According to the company, the solution saved an estimated six months of platform development, and connecting a new API now takes hours instead of days. These results are reported in a case study to which representatives from MoEngage contributed. Details are available in the source article.

What changed

Why it matters

The case study describes a specific architecture for making existing internal functionality available to AI agents. Teams with a larger number of private APIs can use this pattern to assess a shared layer for tools, permissions and connectivity; the reported time savings relate to the deployment at MoEngage.

Two audiences, two different impacts

What this means

01

For individuals

For product API developers, the case illustrates a division of work in which they prepare OpenAPI specifications and select the available operations, while a central team takes on implementation of the MCP layer.

What to do For a selected internal API, verify whether the OpenAPI specification accurately describes the operations you want to make available to an agent.
More practical updates →
02

For a business

Centralization at MoEngage put a single team in charge of managing the MCP layer and agent permissions. According to the company, this eliminated some platform development work and reduced the time needed to connect additional APIs.

Development
What to decide When evaluating a shared MCP layer, verify whether it provides private connectivity and server-side enforcement of permissions for individual agents.
More business impacts →
AgentCore Gateway Amazon Bedrock AgentCore MCP MoEngage OpenAPI

Check the original

Event sources

only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected How MoEngage turned over 400 APIs into agent tools with AgentCore Gateway