Turn Plain English Into Automated Tests with AI Translator
One of the biggest additions in Rapise 9.1 is AI Translator, which lets testers describe automation steps in natural language and have Rapise translate those instructions into executable SmartActions.
Instead of having to express every interaction in automation-specific syntax, a tester can write instructions such as “Log in as the standard user,” “Search for the customer record,” or “Verify that the order appears in the results.” AI Translator interprets the intent of those instructions and generates the automation needed to perform them.
Importantly, AI Translator is not simply generating generic automation code. It works with Rapise's existing testing framework, including Page Objects, Modules, learned objects, and the Object Repository. It translates natural-language instructions into SmartActions, giving the resulting automation access to the same self-healing capabilities that help Rapise tests adapt when applications change.
This can significantly lower the barrier to creating automated tests. Subject-matter experts and manual testers can describe test scenarios using language that is familiar to them, while experienced automation engineers can use AI Translator to accelerate repetitive test development.
It also helps teams preserve an important connection between what a test is intended to do and how it is automated. The natural-language description remains easy for a human to understand, while Rapise handles much of the technical translation required to execute it.
AI That Understands the Context of the Test
AI Translator is designed to work with more than isolated commands. It maintains context across multiple test steps, allowing it to interpret each instruction as part of the larger test scenario.
It can also use information from the Rapise automation framework when determining how a step should be implemented. This includes awareness of Modules, Page Objects, and repository objects, as well as AI-assisted locator generation. The result is AI-assisted automation that is grounded in the actual structure of the test rather than disconnected code generation.
For testing teams, this means natural language can increasingly become another way of authoring automation rather than merely a way of asking an AI assistant for suggestions.
Introducing Agentic MCP Chat
The second major AI capability in Rapise 9.1 is MCP Chat.
MCP Chat introduces an agentic conversational interface that connects AI models with the tools and information available through Model Context Protocol (MCP) servers. In Rapise 9.1, the chat can work with both the Rapise MCP Server and the Spira MCP Server, creating a bridge between test automation and test management.
Instead of navigating through multiple screens or manually invoking individual commands, users can describe what they want to accomplish in normal language.
For example, a tester might ask Rapise to find the tests associated with a particular area of the application, inspect an existing test case, create a new test, execute automation, or investigate the outcome of a test run.
The AI agent determines which available MCP tools it needs to use and can orchestrate multiple tool calls to accomplish the requested task. Rapise can also present the proposed plan before execution, allowing users to review what the agent intends to do before approving actions.
This is an important distinction between conventional AI chat and agentic AI. The assistant is not limited to answering a question based on information contained in its language model. Through MCP, it can interact with the testing environment itself.
Connecting Rapise Automation and Spira Test Management
The ability to connect to both the Rapise and Spira MCP servers makes this especially powerful for organizations using the two products together.
The Rapise MCP Server exposes testing capabilities such as searching and reading tests, executing tests, accessing object repositories, and working with Page Objects and Modules. The Spira MCP Server provides access to the wider test-management and application-lifecycle context, including test cases, test sets, requirements, releases, test runs, and automation hosts.
That means a conversation can increasingly span the boundary between managing a test and automating a test.
A user could begin with a question about the tests associated with a requirement, investigate the existing automation, make changes, execute the test, and then examine the resulting test information. MCP provides the underlying mechanism through which the AI agent can interact with the different systems required to complete that workflow.
This is part of the larger shift we see happening across software development: AI is moving from being a content-generation tool to becoming an active participant in the software delivery lifecycle.
Bring Your Own LLM — or Use Inflectra.ai
Another important differentiator in Rapise 9.1 is choice.
Organizations do not have to adopt a single AI provider in order to take advantage of Rapise's AI capabilities. Rapise supports Inflectra.ai as well as external AI providers, including OpenAI and models hosted through Amazon Bedrock.
For organizations that already have an approved enterprise LLM strategy, this means they can use their own model infrastructure with Rapise rather than being forced into a separate AI ecosystem.
Alternatively, teams can use Inflectra.ai, which provides an integrated option for accessing the AI capabilities in Rapise and Spira without needing to configure and maintain a separate AI provider. Rapise 9.1 specifically adds support for AI Translator and AiRobot through Inflectra.ai.
Inflectra.ai is particularly attractive for organizations that want predictable AI costs. Rather than continually metering individual prompts and tokens, Inflectra.ai is available through flat-rate subscriptions with unlimited token usage, making it easier for teams to deploy AI broadly without worrying that increased adoption will result in unpredictable usage charges.
That becomes increasingly important as AI moves deeper into test automation. When AI is generating test steps, analyzing automation, helping repair tests, and orchestrating agentic workflows throughout the day, a usage model based around constantly monitoring individual token consumption can quickly become cumbersome.
More AI Throughout Rapise
AI Translator and MCP Chat are the headline features, but Rapise 9.1 extends AI into other areas of the product as well.
The release introduces an RVL AI Assistant, which allows users to describe changes they want to make to an RVL sheet using natural language. Rapise generates a proposed RVL patch that can then be reviewed before being applied. The release also adds in-place editing for AI commands and additional integration between AiRobot and SmartAction recovery.
Rapise 9.1 also includes a broader collection of automation improvements, including new controls for self-healing, support for :visible and :hidden CSS pseudo-classes, enhancements to Page Objects and Libraries, and support for longer Windows file paths.
These enhancements continue the work introduced in Rapise 9.0 around intelligent, resilient test automation while making AI a more integrated part of the overall experience.
From Codeless Automation to Intent-Driven Automation
For many years, one of the goals of test automation tools has been to reduce the amount of specialized coding required to automate a test.
AI changes that equation again.
The question is increasingly no longer simply whether a test is coded or codeless. The more interesting question is whether the tester can describe what they want to accomplish, and allow the automation platform to determine how best to accomplish it.
AI Translator is an important step in that direction. Testers express their intent in natural language, while Rapise translates that intent into automation built on SmartActions and the existing Rapise framework.
MCP Chat takes the idea further. Instead of translating an individual test instruction, an AI agent can understand a higher-level objective, select the appropriate tools, and perform a sequence of actions across Rapise and Spira.
The result is a progression from test automation toward AI-assisted test engineering.
Available Now in Rapise 9.1
Rapise 9.1 is available now and includes AI Translator, agentic MCP Chat, the RVL AI Assistant, expanded Inflectra.ai support, and numerous improvements across the Rapise automation platform.
With this release, we are making it possible for testers to work with automation in a much more natural way: describing tests in plain English, collaborating with AI inside their testing environment, and using agentic workflows that can interact directly with both automation and test-management systems.
And we believe this is only the beginning.
As AI becomes increasingly capable of understanding intent, accessing project context, and taking actions through technologies such as MCP, the role of the testing platform changes as well. Instead of simply giving testers tools that they need to operate, platforms such as Rapise can increasingly help testers decide what needs to be done — and help them do it.
