Despite recent methods As recent technologies emerge, enterprises continue to turn to autonomous coding agents and code generation platforms. Competition to keep developers working on their platforms from technology companies has also intensified.
AWS believes its offer Kiraand recent capabilities for behavioral compliance are a massive differentiator in an increasingly crowded coding agent space.
Kira, first launched in July in public preview, is now generally available with recent features including property-based behavior testing and a command-line interface (CLI) for customizing custom agents.
Deepak Singh, AWS’ vice president of databases and artificial intelligence, told VentureBeat in an interview that Kiro “keeps the fun” in coding while providing structure to it.
“The way I like to say it is that Kiro allows you to talk to an agent and work with them on software development, just like you would with any other agent,” Singh said. “But what Kiro does is that it takes this structured way of writing software, which we call spectrum and development, down to specifications that take your ideas and turn them into things that will last over time. The result is more robust and maintainable code.”
Kiro is an agent coding tool built into developer IDEs that helps build agents and applications from prototype to production.
In addition to the recent features, AWS is offering startups in most countries a year of free credit for Kiro Pro+ and expanded access to Teams.
Built-in compliance and checkpoints
One of Kiro’s recent features is property-based testing and checkpointing.
The problem some companies face with AI-generated code is that it is sometimes challenging to assess the accuracy and degree to which agents are achieving their intended goals. AWS noted in a blog post that “whoever writes the tests (human or AI) is limited by their own biases – they have to think about different, specific code testing scenarios, and they will miss edge cases they haven’t thought about. AI models often ‘cheat’ the solution by modifying tests rather than fixing the code.”
“Property-based testing is what it does, it takes the specification, it takes the specification, and based on that it identifies the properties that your code should have, and it basically creates potentially hundreds of test scenarios to verify that your code does what it intended to do, as stated in the specification, and it does it all automatically,” Singh said.
Singh said organizations can submit their specifications and the Kiro agent can start identifying what’s missing even before the code review process begins.
Property-based testing matches specific behavior, or statements, to what the code does. Kiro can aid users write this into their specifications based on the EARS format. For example, if a company is creating a car sales app, the specification would be:
“For any user and any list of cars, WHEN the user adds a car to favorites, the system MUST display that car in the favorites list. PBT then automatically tests this using user A adding car #1, user B adding car #500, user C adding multiple cars, users with special characters in their usernames, cars with different statuses (recent, used, certified), and hundreds of other combinations, catching the latest cases and verifying that the implementation matches yours intentions.
Unlike the conventional unit test specification which states: If the user adds car #5 to their favorites, it will appear in their list.
Kiro will then identify examples of code that violate the specifications and present them to the user.
Kiro now also allows you to create checkpoints so developers can revert to a previous change if something goes wrong.
CLI coding
The second major recent feature of Kiro is the Kiro CLI, which brings the Kiro coding agent directly to the developer’s CLI.
AWS claims that Kiro CLI uses some functionality from Q Developer CLI – a built-in coding assistant, launched in October 2024—to allow users to access the agent from the command line.
It also enables developers to start building custom agents, such as Backend Specialist, Frontend Agent, and DevOps Agent, tailored to the organization’s codebase.
Singh said developers have their own unique ways of working, so it’s crucial for coding agent providers like AWS to meet them where they are. Kiro CLI enables users to:
-
Stay in the terminal without having to switch contexts
-
Structuring AI workflows with custom agents
-
It needs to be configured for two environments because MCP servers and other tools run both the IDE and CLI versions of Kiro
-
Quickly automate code formatting or log management with automated commands
Coding agent competition
However, Kiro is just one of many coding agent platforms emerging and competing for utilize in enterprises.
WITH “OpenAI”s GPT-Codex, which unifies the Codex coding assistant with IDE, CLI and other workflows to “Google”with Gemini CLI, it’s clear that more and more developers require simple access to coding agents where they do their work.
And enterprises demand more from coding agents. For example, Anthropic it worked Claude Code platform available on the Internet and on mobile devices. Some encoding platforms also allow users to choose the model they will utilize for encoding.
Singh said Kiro is not based on just one LLM; instead, it directs you to the best working model, including AWS models. At launch in July, Kiro was based on Claude Sonnet 3.7 and 4.0.
Well-known brands, e.g Monday.com To have noticed significant benefits AI-based coding, showing that enterprises will likely continue to utilize these platforms in the future.
“What we’ve seen is that developers’ mental model is changing, but it’s not just about being more efficient, it’s also about how they now organize themselves based on the way they work,” Singh said.