Release Notes: testRigor MCP Server, Better AI Tests, Rename Variables
 |
testRigor MCP Server
We are excited to introduce a new feature in testRigor: a customer-facing MCP server that enables seamless integration with agentic AI tools such as Claude and ChatGPT. This integration allows users to interact with testRigor using natural language, making it possible to create test cases, execute tests, and manage automation workflows directly through AI assistants.
The setup process is straightforward. Users generate a Personal Access Token (PAT) within testRigor and then configure the MCP server inside their chosen AI tool. This connection establishes secure communication between testRigor and the AI environment. Step-by-step setup instructions are available in our recent blog post: https://testrigor.com/how-to-utilise-testrigors-mcp-server/
With this release, users gain access to a wide range of capabilities through AI-driven commands. These include listing test suites and test cases, executing individual tests, triggering full test suite runs, retrieving execution results, identifying failures within a run, and canceling ongoing tasks. A complete overview of supported actions is available in the setup guide.
This feature represents a significant step forward in bringing test automation and agentic AI together. By combining testRigor’s low-maintenance, plain English test automation with the reasoning and orchestration capabilities of modern AI tools, teams can accelerate testing workflows, reduce manual effort, and improve overall productivity.
Better AI Generated Tests
We added a new capability to AI-generated test cases by using vector search to find similar example tests. The AI locates successful test examples and reuses relevant steps and reusable rules, such as “search and select a sweater,” as templates to generate new test cases faster and with fewer interactions. The workflow now extracts multiple steps from a matched scenario and applies them to a new task, for example changing “search for a shirt” to “search for a t-shirt” while preserving the same step pattern. The AI executes those grouped steps in a single interaction, collects the results, and proceeds to generate the full test case. Because the AI uses vector search to find closely matching scenarios, it borrows proven steps and produces higher-quality test cases with less back-and-forth. The result is faster test-case generation, fewer AI interactions per test, improved quality through reuse of successful rules, and easier maintenance as templates adapt to small variations.
Rename Global Variables
You can now rename test data variables globally across the test suite. In the test data page there is a new Rename button next to the variable name. Clicking it updates the name, displays a confirmation alert that the rename succeeded, and immediately updates every reference to that variable in test cases and reusable rules. This reduces manual edits and prevents inconsistencies when variable names change.
