In our third Kong and Okta tutorial, we'll go through the introspection flow implementation. This series will show you how to implement service authentication and authorization for Kong Konnect and Okta using the OpenID Connect (OIDC) plugin. Parts 1, 2 and 4 cover:
In this example, I'm using the Konnect control plane to create new APIs and policies and publish them to my data plane running as a Docker container in an AWS EC2 instance.
The introspection flow is part of the token validation process. Kong Gateway evaluates the injected token at the request processing time to see if it’s still valid to the upstream services. The evaluation hits a specific Okta endpoint, passing the received token. Based on the response provided by Okta, Kong Gateway accepts or rejects the request.
For production environments, the OIDC plugin provides caching capabilities for the Okta responses. However, for this tutorial, I’m going to disable caching to better view the flow.
Regarding Okta’s settings, I’m going to use the same client credentials application I created before. With the client ID and client secret. However, my OIDC plugin has to be set with specific parameters to implement introspection.
In the Konnect ServiceHub, I have an IntrospectionRoute OIDC plugin enabled.
The settings should be:
To better view the flow, I will use Insomnia, Kong’s API spec editor, to send requests to both Okta and Konnect. Below are my two requests.
The first one I’m sending to Okta, passing the expected parameters to get it authenticated and receive a token.
For the second one to consume the route, I’m using a specific Insomnia capability called Request Chaining. With this, I’ll be able to extract values from the response of a given request to build new ones. In my case, I’m pulling the access token from Okta’s response to make the other request and then send it to Konnect.
Next, let’s send a request to Okta to get our token. There it is.
This time, we can see that Kong’s request is ready to be sent since we got Okta’s token injected inside of it.
And here’s the Konnect response:
It's important to note that Konnect is validating the token behind the scenes. Here's one EC2 terminal where my data plane is running. Since I disabled introspection caching for the OIDC plugin, Konnect hits Okta for each request to validate the token.
Another way to see introspection is by deactivating the Okta application. All tokens related to it will be considered invalid and, as a consequence, will not be accepted by Kong again.
Let’s get back to Okta’s application and deactivate it. We should get a 401 error code from Kong.
Start a free trial, or contact us if you have any questions as you're getting set up.
Once you've set up Konnect and Okta introspection flow, you may find these other tutorials helpful:
Meet Emily. She’s an API product manager at ACME, Inc., an ecommerce company that runs on dozens of APIs. One morning, her team lead asks a simple question: “Who’s our top API consumer, and which of your APIs are causing the most issues right now?”
Ever feel like you're fighting an uphill battle with your API strategy? You're building APIs faster than ever, but somehow everything feels harder. Wasn’t API-first supposed to make all this easier? Well, you're not alone. And now industry analys
Kong-quer the Agentic AI Hackathon 🚀 Calling all builders, tinkerers, and API innovators. The Kong Hackathon is back for API Summit 2025 ! This year, we’re challenging developers worldwide to create projects that don’t just react, they think , a
In the last two parts of this series, we discussed How to Strengthen a ReAct AI Agent with Kong AI Gateway and How to Build a Single-LLM AI Agent with Kong AI Gateway and LangGraph . In this third and final part, we're going to evolve the AI Agen
In my previous post, we discussed how we can implement a basic AI Agent with Kong AI Gateway. In part two of this series, we're going to review LangGraph fundamentals, rewrite the AI Agent and explore how Kong AI Gateway can be used to protect an LLM
This is part one of a series exploring how Kong AI Gateway can be used in an AI Agent development with LangGraph. The series comprises three parts: Basic ReAct AI Agent with Kong AI Gateway Single LLM ReAct AI Agent with Kong AI Gateway and LangGr
What Is RAG, and Why Should You Use It? RAG (Retrieval-Augmented Generation) is not a new concept in AI, and unsurprisingly, when talking to companies, everyone seems to have their own interpretation of how to implement it. So, let’s start with a r