Page History

This release includes the following enhancements.

Release date: Month Date, Year

AI-Related Tools and Improvements

Several AI-related tools and improvements are included with this release. Each of these requires an LLM Integration license. Additionally, you must configure your LLM provider account information in the LLM Provider preferences. OpenAI, Azure OpenAI, and other LLM providers with similar functionality are supported. See LLM Provider preferences to determine whether your LLM provider can be used.

AI Assertor Tool

Testing AI-Infused Applications

This release includes two new tools designed to make testing AI-infused applications easier and more reliable: the AI Assertor and the AI Data Bank. In addition, you can now also use SOAtest as a Model Context Protocol (MCP) server so you can integrate it with AI-powered code editing tools like Copilot and Windsurf.

AI Assertor

The AI Assertor The AI Assertor tool is an LLM-powered message analyzer that uses your LLM provider to determine whether a message satisfies one or more assertions or requirements as written compares messages to your requirements—written by you using natural language. It language—to make sure they are met, even when phrasing differs. This makes it an invaluable tool for validating conversational outputs and dynamic content from LLMs without requiring exact matches. The AI Assertor is most commonly connected to a tool that sends or receives messages in order to verify the content of those messages. See See AI Assertor for more information.

AI Data Bank

When you need to extract and reuse data between test steps, like capturing a name, balance, or reference number, the AI Data Bank lets you define the extraction logic in natural language. It identifies the right data from previous responses and passes it forward automatically, eliminating the need for hard-coded or complex definitions of what to extract. It

The AI Data Bank tool is an LLM-powered message analyzer that uses your LLM provider to power data banking that easily adapts to changing conditions by using natural language to define what conditions you want to look for. It is most commonly connected to a tool that sends or receives messages in order to extract data from those messages for use by another tool in the test scenario. See AI Data Bank See AI Data Bank for more information.

Testing and Simulation of MCP Servers

An MCP server acts as the connection point between the LLM and the external tools it can call. When the model decides to use a tool, it sends a request to the MCP server, which then handles the tool invocation and returns the result. You can configure AI-powered code editing tools like Copilot and Windsurf to use SOAtest as an MCP server and, once configured, those tools can use SOAtest directly. This works with both the SOAVirt WAR and desktop server. This feature requires a license that has "LLM Integration" enabled and is still in beta. See Configuring SOAtest as an MCP Server for more information.

MCP Client Tool

SOAtest now has an MCP Client tool, enabling you to take full advantage of SOAtest’s rich interface when configuring, sending, and validating messages sent over MCP. This extension can be used to test a MCP server and supports messages sent over HTTP using Server-Sent Events (SSE). See MCP Client for more information.

Creating Assets from the AI Assistant

The AI Assistant has been improved to be able to help you with even more tasks, including:

   — Creating virtual assets, either with or without a service definition, with parameterization. See Creating Virtual Assets with the AI Assistant and Parameterizing Tools with the AI Assistant for more information.

   — Applying parameterization to tests it helps you create. See Creating Tests with the AI Assistant#Parameterize for more information.

   — Creating a message proxy for data learning. See Setting Up a Data Learning Environment for more information.

LLM Provider Settings for WAR Deployments

You can now configure LLM provider settings for WAR deployments so you can use AI-related tools with the SOAtest and Virtualize server. See Configuring Virtualize Server#LLM Provider Settings for more information. See also Configuring Settings#LLM Provider Settings for more information regarding setting up your LLM provider in your settings configuration file.

You can configure AI-powered code editing tools like Copilot and Windsurf to use SOAtest as a Model Context Protocol (MCP) server. Once configured, those tools can use SOAtest directly. This works with both the SOAVirt WAR and desktop server. This feature requires a license that has "LLM Integration" enabled and is still in beta. See Configuring SOAtest as an MCP Server for more information.

MCP Client Tool

SOAtest now has an MCP Client tool, enabling you to take full advantage of SOAtest’s rich interface when configuring, sending, and validating messages sent over MCP. This extension can be used to test a MCP server and supports messages sent over HTTP using Server-Sent Events (SSE). See MCP Client for more information.

Manual Testing Coverage Improvements in CTP

The Coverage Agents Table, which allows you to review all of your components and which coverage agents they are connected to in a convenient table, has been moved to the Environments page. See Connecting CTP to Coverage Agents#Coverage Agents Table for more information. Now when you select an environment in the Manual Testing module, you will see a table with the most recent manual testing results for the environment, including information for each test like when it was last tested and by whom. See Manual Tests for more information.

In addition, you can now create different versions of your CTP systems to better support your development cycle. This gives you the ability to have a version that represents, for example, your baseline system plus additional versions for different releases. You can configure these versions to include some or all of the components and environments from the version they're based on, including whether to use shared components or standalone duplicates. Any number of versions can be defined for a system. See Defining Versions for more information.

Additional Updates

• CTP now supports PostgreSQL 17.
  

  Warning
  If you want to migrate your current CTP database to PostgreSQL, you must update CTP to 2025.2 before you back up your existing database. Only then will you be able to successfully migrate your CTP database to PostgreSQL.

• Added "ProvisioningAssets" to Workspace Files on remote server to make it possible to upload pvn files from the desktop to remote servers.
• Added "Execute on Server" action to Server View so that users can run a single tst/pvn on a remote server.
• Added a Demo Configuration to the list of configurations that ship with the SOAtest and Virtualize server. See Test Configuration Options for SOAtest and Virtualize Server for more information.

• GraphQL users can now configure the GraphQL Message Responder so that, when it is constrained to an SDL, it will return only the fields requested by the client, even if additional fields are specified in the Response tab.

Resolved PRs and FRs