Hi! I’m Girish, an AWS Community Builder and Cloud Tech Enthusiast, with expertise in delivering customer-focused and business-impacting cloud transformation programs of high complexity.

In my previous articles, I shared various Kiro features like vibe coding, specs driven development, steering docs and hooks. All these features were available prior to reinvent 2025.

This time, I am writing about a feature of Kiro that was announced during reinvent 2025. This feature is Kiro Powers!

I will use Postman Power in Kiro to demonstrate how you can use agentic AI to create API test collection and then use it in Postman to validate your API endpoints.

What Did I Build and Why?

A few weeks back, I built a CRUD API for customer look up using AWS Kiro. Based on my prompt, Kiro created multiple API endpoints to lookup, create, update and delete customer information. It created a tech stack including API Gateway, Lambda and Dynamo DB. An IaC code using AWS SAM was created as well to build and deploy this solution to AWS Cloud. At that time, I used Postman to validate the API endpoints and created the needed postman collection manually.

Fast Forward to December, post reinvent 2025, now Kiro supports Powers and one of the powers is Postman Power for which Postman, the creator of the tool, is the creator of this power.

I decided to try out this Postman power and used the same API endpoints to validate this. With this Postman power in action, Kiro created the entire collection in minutes.

This wasn’t just a prototype or proof of concept, it was an opportunity to see how Kiro’s AI-driven coding capabilities could help me automate the repetitive tasks of API validation collection creation, thus saving the time and allowing me to focus on ensuring that APIs are built correctly and to the specs.

Architecture

What Are Kiro Powers

Kiro Powers are modules created in partnership with developer tool providers to supply tool-specific context and expertise to agentic AI–based development workflows. In simple terms, Powers combine tool-specific MCPs and steering documentation into a single unit that guides the agentic AI with the right context resulting in more accurate responses and higher-quality code or artifact generation.
There are many different Powers available today, including Postman, Figma, Terraform, Stripe, and more.

In this article, I’ll be using the Postman Power with Kiro to demonstrate how it can generate a complete API collection.

How Kiro and I Did It!

I started by providing Kiro a prompt.

Here’s the exact prompt I used:

“I have a customer lookup API. Using Postman Powers, create a collection for getCustomer API.”

That’s it. No schema definitions. No OpenAPI spec. No manual instructions.

What I found interesting that instead of blindly creating something new, Kiro responded by evaluating my existing workspace. As part of this assessment, it found the Customer Lookup API and confirmed that it will be building the API collection for this Customer Lookup API.

Full CRUD Collection in Minutes

Kiro recognized that my Customer Lookup API includes multiple endpoints such as Get Customer, List Customers, Create Customer, Update Customer, and Delete Customer. It automatically created a Postman API collection for these endpoints, ensuring that each request includes the appropriate parameters and is ready to run.
Here are the API endpoints Kiro created the collection for:

• Get Customer – Retrieve a customer by ID
• List All Customers – Retrieve all customers
• Create Customer – Add a new customer
• Update Customer – Modify an existing customer
• Delete Customer – Remove a customer

Kiro, not only created the collection but also included Environment variables for endpoint portability, it also added API key–based authentication headers.

Testing the Collection with Postman

Once Kiro confirmed that the collection had been created, I wanted to test it using Postman.

However, I couldn’t initially find the collection in my local workspace. Since Kiro was acting as my peer programmer, I decided to ask where the collection had been saved.

As shown in the screenshot below, Kiro responded just like a true peer programmer, explaining that the collection was saved directly in my Postman environment. This eliminated the extra step of manually importing the collection into Postman and allowed me to start testing immediately.

View from the Postman

Here is the view in my Postman environment. I can see this collection in there, it was added by Kiro.

Get Customer Validation

This is the result from Get Customer API invoked using the collection.

![Image getcust(https://dev-to-uploads.s3.amazonaws.com/uploads/articles/4wpqq3inzh4mabhcb7il.png)

Create Customer Validation

This is the result from Create Customer API invoked using the collection.

When I run the Get Customer again, I get the updated list including the one customer, I just created.

Conclusion

In this article, I demonstrated how Kiro Powers, combined with Postman Powers, can accelerate API development and testing by eliminating repetitive setup work. Instead of spending hours manually creating requests and configuring environments, Kiro evaluated my existing workspace and generated a complete Customer API collection with all required endpoints ready to use.

The real magic was in the feedback loop. Kiro didn’t just create requests—it understood the context of the existing API, generated a well-structured Postman collection, and clearly explained what was created and why. It felt less like a tool and more like a technical partner supporting the API testing workflow.

What started as a simple request to generate a Postman collection quickly became a practical demonstration of how agentic AI can improve everyday developer workflows. With automated tests, environment variables, authentication headers, and validation already in place, the result was a fully usable Postman collection that could be tested immediately.

Whether you’re building a new API, refining an existing one, or standardizing collections across teams, Kiro’s blend of AI-assisted automation and contextual understanding makes it easier than ever to move from idea to tested implementation—fast, accurate, and reliable.

Generative AI isn’t replacing good API design or testing practices; it’s enhancing them. And with tools like Kiro, even routine tasks like creating Postman collections can feel like working alongside an experienced API engineer who understands your workspace.

Generative AI is transforming how APIs are built and tested, and tools like Kiro are making that transformation practical. While increased AI-generated artifacts can introduce new challenges, features like context awareness, spec-driven workflows, and automated validation help maintain quality and consistency.

I believe this is just the beginning, and these tools will continue to evolve rapidly.

Thanks for reading, and I hope you found this insightful.

Watch the video here:

Thanks,
𝒢𝒾𝓇𝒾𝓈𝒽 ℬ𝒽𝒶𝓉𝒾𝒶
𝘈𝘞𝘚 𝘊𝘦𝘳𝘵𝘪𝘧𝘪𝘦𝘥 𝘚𝘰𝘭𝘶𝘵𝘪𝘰𝘯 𝘈𝘳𝘤𝘩𝘪𝘵𝘦𝘤𝘵
𝘈𝘞𝘚 𝘊𝘦𝘳𝘵𝘪𝘧𝘪𝘦𝘥 𝘋𝘦𝘷𝘦𝘭𝘰𝘱𝘦𝘳 𝘈𝘴𝘴𝘰𝘤𝘪𝘢𝘵𝘦
𝘈𝘞𝘚 𝘊𝘦𝘳𝘵𝘪𝘧𝘪𝘦𝘥 𝘎𝘦𝘯𝘈𝘐 𝘗𝘳𝘢𝘤𝘵𝘪𝘵𝘪𝘰𝘯𝘦𝘳
𝘈𝘞𝘚 𝘊𝘭𝘰𝘶𝘥 𝘛𝘦𝘤𝘩𝘯𝘰𝘭𝘰𝘨𝘺 𝘌𝘯𝘵𝘩𝘶𝘴𝘪𝘢𝘴𝘵