Introducing API Agent: Democratizing APIs in an Agentic Era

Today at THINK, IBM API Connect is announcing the new API Agent—built to accelerate your API journey with greater efficiency, speed, and confidence.

AI is no longer just a futuristic concept; it’s reshaping not only what we build, but how we build. The rapid evolution of large language models (LLMs) and AI agents is driving seismic shifts in the world of API consumption and creation APIs have moved beyond simply providing access to data. They now make it easier and faster for developers, businesses, and AI agents to build, integrate and innovate.

The way that we interact with businesses is changing, and those who do not embrace this change will be left behind Instead of interacting directly with websites and mobile applications, people are communicating with AI models, who in turn are dispatching AI agents to execute tasks on the user’s behalf If your API isn’t AI-ready, then it might as well not exist To succeed, you need an API landscape where users and agents can bring ideas to life with minimal friction.

In this new model, AI boosts the productivity of API teams by helping them maintain flow state and automates routine tasks along the API lifecycle. But Agentic AI takes this even further, introducing agents that can both create new APIs and manipulate your IT environment to test and deploy them

At the same time, you can build an AI Agent to book a flight, track a package, rent a car or perform any other of your core business functions that you’ve programmatically exposed. This creates a self-reinforcing feedback loop, and the net impact will be accelerated change.

While Agentic AI is a transformative technology that is changing the way we interact with the world, the increase in velocity has exacerbated challenges around sprawl, governance and security, making it crucial to have a comprehensive API strategy for the agentic era.

• API sprawl: The surge in new APIs—many auto-generated for AI use-cases or existing in silos across different environments—increases operational complexity and risk.

• Increased security vulnerabilities: Every exposed API is a potential attack vector if it’s not secured with strong authentication, authorization and threat protection. As AI agents autonomously discover and consume APIs, the sheer volume of API calls multiplies, making it more difficult to identify malicious actors.

• Need for governance: Without intentional governance, enterprises risk having APIs that are poorly designed and difficult to consume.

Developers will be tempted to create new APIs instead of re-using existing ones, creating a vicious circle of sprawl, vulnerabilities, and technical debt.

Our new API Agent has been trained on the API Connect platform APIs and has access to your enterprise resource catalog, so it both understands your API environment and how to use API Connect to prevent these problems from occurring. It responds to requests given in plain English, using its knowledge of software engineering, API best practices, API Connect capabilities and your API estate to turn your intent into reality.

Consider a scenario where Acme Coffee Roasters is expanding their delivery internationally and partnering with Just in Time logistics. They want Just in Time to query order details and push shipping-status updates. Acme’s developer tells the API Agent, “Create an API to retrieve order details and update shipping statuses in real time.”

Before generating a brand-new API, API Agent intelligently searches the existing API estate to discover if a suitable API already exists—promoting reuse and minimizing sprawl. If no suitable API is found, the Agent automatically generates the API specification. It supports both code-first and API design-first approaches, recognizing that different teams have different workflows. Some prefer to design and document their APIs first (design-first), while others want to focus on building backend functionality first (code-first), then generate the API specification based on the implemented code to speed up prototyping.

By embracing both models, API Agent adapts to real-world development lifecycles, meeting teams where they are and removing barriers to faster, safer API delivery. Additionally, it augments the specification with relevant documentation providing richer information and context for the usage of APIs by applications. API Agent can also ensure the APIs adhere to organizational standards and best practices by validating them against governance rulesets. It can also catch and fix validation errors early in the development process. Finally, it generates re-usable test cases to ensure the API behavior is valid over time.

As the boundaries between humans, systems and AI continue to blur, the ability to seamlessly create, manage and scale APIs will be a competitive advantage. With API Agent, enterprises can stay ahead—building smarter, governing better and unlocking the full potential of an AI-driven, agentic future.

Read more about how IBM webMethods Hybrid Integration and API Agent together streamline integration and accelerate automation.