From Spreadsheets to BDD: How to Centralize Test Scenarios

Moving from spreadsheets to BDD test management means your scenarios live in one searchable place, with execution history and collaboration built in. This guide walks you through the migration in five steps, then shows how you can automate both the move and ongoing test runs using an AI assistant and the Gwirian skill.

Why move off spreadsheets?

Spreadsheets work for a handful of test cases. Once you have many scenarios, multiple people, or a need for traceability (who ran what, when, pass or fail), they break down: no full-text search, no execution history, no clear ownership. A dedicated BDD test management tool gives you one place to create scenarios, track runs, and search across projects. The table below sums up the difference.

Part 1: Migrate in five steps

1. Export and audit what you have

Export your test cases from Excel, Google Sheets, or wherever they live (CSV is fine). List everything you actually use: scenario name, steps, expected result, maybe a “feature” or “module” column. Drop duplicates and obsolete cases. This audit is the basis for your BDD structure.

2. Choose a structure: features and scenarios

In BDD you group scenarios under features. One feature might be “User login” or “Checkout flow”; under it you have scenarios like “Valid login” or “Login with wrong password.” Map your current rows or tabs to features, then to scenarios. If you don’t have a feature column, group by area (e.g. “Auth,” “Payments”) and name the feature from that.

3. Write Given-When-Then

Turn each scenario into Given-When-Then steps. For example:

• Given the user is on the login page
• When they enter valid credentials and submit
• Then they see the dashboard

This format is standard (Gherkin) and works in any BDD tool. You don’t have to convert everything in one go; start with the most important flows.

4. Pick a tool and create the structure

Create a project (and features) in the tool you chose. Gwirian is one option: create a free account, create a project, then add features and scenarios. You can type them in the UI or, as in Part 2, use the CLI and an AI assistant to create them from a spreadsheet or CSV.

5. Run and track execution

Once scenarios are in the tool, run them (manually or via automation) and record pass/fail. Over time you get a single place to see what’s been run, by whom, and when—no more hunting through sheets or emails.

Part 2: Automate with AI and the Gwirian skill

You can speed up the migration and ongoing test runs by using an AI assistant (Cursor or Claude Code) that knows how to talk to Gwirian via the gwirian-cli and the Gwirian skill.

What is the Gwirian skill?

The Gwirian skill is a set of instructions you install so that Cursor or Claude Code can use gwirian-cli correctly. The CLI talks to the Gwirian API: list projects, features, and scenarios; create or update features and scenarios; and record scenario executions. The skill teaches the AI when and how to run those commands (for example, to create scenarios from a CSV or to record a test run after Playwright executes a scenario).

Install the CLI and the skill:

npm install -g @acmada/gwirian-cli
gwirian auth
gwirian install --skills

Use gwirian install --skills --global to install the skill in your home directory so all projects can use it. After that, your AI assistant can run gwirian features list, gwirian scenarios create, and similar commands when you ask it to work with Gwirian.

Use AI to create scenarios from a spreadsheet or CSV

Instead of retyping scenarios, you can hand the AI a CSV or a paste of your spreadsheet and ask it to create features and scenarios in Gwirian. For example:

• “I have a CSV with columns: feature, scenario_title, given, when, then. Create a Gwirian project (or use project ID X), create a feature for each unique feature name, and create scenarios under each feature with the given/when/then steps.”

The AI uses the Gwirian skill to run gwirian features create and gwirian scenarios create with the right project and feature IDs. You get your scenarios in Gwirian without manual data entry.

Use AI to run scenarios and record results

After scenarios live in Gwirian, the same AI and skill can run them (for example with Playwright in the browser) and record pass/fail back to Gwirian. The flow is: AI fetches project and scenario data via the CLI, uses project context (URLs, logins) to run the test, then records the scenario execution with gwirian scenario-executions create. For the full setup—gwirian-cli, Playwright CLI, and how to prompt the AI—see Automate BDD Tests with Gwirian, Playwright, and a LLM.

Prerequisites for the AI workflow

• A Gwirian account and an API token (from the hosted app or your self-hosted instance)
• Node.js 18 or later
• gwirian-cli installed and configured (gwirian auth)
• The Gwirian skill installed in Cursor or Claude Code (gwirian install --skills)
• For browser automation: Playwright CLI and the workflow described in the automation post

Next steps

1. Export and audit your current test cases from spreadsheets.
2. Map them to features and Given-When-Then scenarios.
3. Create a Gwirian project and add features/scenarios (by hand or via the CLI and AI).
4. Install gwirian-cli and the skill so your AI assistant can create scenarios and record runs.
5. Optionally follow the Playwright + LLM automation guide to run scenarios in the browser and record results back to Gwirian.

Try Gwirian free—no credit card required. Self-host for free or use the hosted app to centralize your BDD scenarios and execution history.