Apollo GraphQL, the graph-based API orchestration leader, is expanding GraphQL as the essential orchestration layer for AI agents and modern applications, introducing enhanced AI tooling with Apollo MCP Server 1.0, GraphOS MCP tools to support agentic graph development, Apollo GraphOS Operator for Kubernetes, and enterprise platform enhancements.
According to the company, these updates position Apollo as the definitive solution for organizations looking to create secure, and scalable connections between AI and real-world APIs.
Apollo is establishing the platform layer that enables any organization to safely and efficiently expose APIs to AI agents and leverage agentic development tools across their API development lifecycle, the company said.
"Adapting APIs for agentic use is an urgent priority for every company; getting it right and keeping up with a fast-changing AI landscape can be tricky and time-consuming,” said Matt DeBergalis, CEO and co-founder of Apollo GraphQL. “The better approach is AI-to-API integration infrastructure that’s secure, observable, and performant by default. At Apollo, we’ve spent a decade making API orchestration reliable at trillion-operation scale. Now we're applying that same rigor to the AI problem: giving agents safe, auditable access to real systems without requiring teams to become AI infrastructure experts.”
The announcements center on immediately available capabilities that demonstrate Apollo's comprehensive approach to AI-application integration across three core pillars:
Connect AI Agents
- Apollo MCP Server 1.0 delivers production-ready AI integration with built-in monitoring to track the tools and API calls within AI agent requests. The visibility, plus the secure access to GraphQL APIs without custom coding, enables teams to experiment and build quickly with confidence.
Enable Agentic API Development
- GraphOS MCP tools, now available in Apollo’s GraphOS platform, enable rapid agentic development of new and existing APIs. These tools can be used in conjunction with a wide range of coding agents such as Claude Code and Cursor to automate large parts of the development process.
- Apollo Connectors Test Framework, Type-to-Run, and other enhancements (in Preview) help teams bring REST APIs into GraphQL confidently with improved testing and local development capabilities. The Apollo Connectors Test Framework enables unit testing directly or via a CI script. The Type-to-Run capability simplifies local development by allowing teams to run Connectors with minimal configuration and setup. Additional Connectors arrow functions provide greater expression and flexibility to bring enterprise REST APIs to the graph.
Tools for Enterprise-grade API Operations
- Apollo GraphOS Operator for Kubernetes declaratively deploys and manages GraphQL environments at scale. Delivering out-of-the-box automation that platform teams have traditionally built from scratch, the Operator streamlines GraphOS Runtime operations, automates schema publishing, composition, deployment and monitoring, so teams can spend less time managing infrastructure and more time shipping new features. Future capabilities will include canary deployments, self-healing router instances, and support for managing Apollo MCP Server to seamlessly connect AI agents to runtime environments.
- Improved GraphOS permissions through API keys which can be scoped to a single microservice, also known as a subgraph, enabling greater security as engineering teams contribute independently to a central, federated GraphQL API.
- Connector and Subgraph insights provides service developers with finer grained insights into their specific service’s behavior, allowing teams to proactively optimize for service performance and scaling, and enabling organizations to better distribute operational responsibilities for a Federated graph.
“What we’re finding is that agents need self-service platform tooling just as much, if not more, than human engineers do, and agents need APIs just as much as the last era of applications” said Rob Brazier, VP of product at Apollo GraphQL. “We want to enable every organization to respond to the AI opportunity, by providing tools that enable every engineer to build, ship and operate APIs more effectively, and enable AI agents to get the context data they need to deliver the promised productivity and experience benefits.”
Apollo Client 4.0 delivers improvements in bundle size, TypeScript support, and developer experience, while Apollo iOS 2.0 introduces performance improvements and modern Swift concurrency features for native mobile applications.
Apollo MCP Server 1.0, Apollo GraphOS Operator for Kubernetes, GraphOS MCP tools, Apollo Client 4.0, and Apollo iOS 2.0 are generally available now, with additional features in preview.
For more information about this news, visit www.apollographql.com.