The Thunders MCP server connects the AI tools your team already uses (Claude, ChatGPT, Cursor, VS Code, Windsurf, Devin) directly to your Thunders workspace.
Author test cases, trigger runs, and diagnose failures in plain language, without leaving the tool you are already working in. No scripts, no API keys, no context switching.
Connect in under 60 seconds
Add a custom connector in your AI client.
Paste the server URL:
https://api.thunders.ai/v1/mcpSign in to Thunders and click Authorize.
What you need
A Thunders account
An MCP-compatible AI client
Where to go next
Set up your client:
Then put it to work.
**Use Cases: how to make the most out of Thunders MCP Server** for prompts you can copy, grouped by the job you are doing
**Thunders MCP available tools** for everything the server can do
What you can do with it
Migrate legacy suites | Turn Playwright, Selenium or Cypress tests into executable Thunders test cases, without rewriting them by hand. |
Author from context | Build test cases from a ticket, a spec, a design file or a URL, then organise them into labelled sets. |
Run with a persona | Execute on the environment you choose, as the persona you choose, in parallel or in sequence. |
Diagnose failures | Pull the full context of a failed run (steps, logs, bug reports) and get a fix proposed against the actual step that broke. |
Analyse coverage | Ask for fail rates, flaky tests, gaps and recurring patterns across your runs. |