Getting started
Create a workspace and a project, invite your team, and find your way around.
sqale QA is where your QA team keeps everything in one place: requirements, test cases, test runs, bugs, releases, CI results and API tests. An AI agent can search and act on all of it.
1. Create your workspace
Sign up at sqalehq.ai. You can work in a personal workspace or create a team workspace (an organisation) for your company. Team workspaces let you invite colleagues and share projects. You can switch between workspaces at any time from the switcher in the top bar.
2. Create a project
A project usually maps to one product or application. Go to Projects → New project, give it a name and a short key (for example SHOP). The key prefixes everything in the project, so requirements become SHOP-12 and test cases TC-34.
Want to look around first? On an empty Projects page, choose Load sample project. You get a realistic e-commerce project with requirements, test cases, runs and bugs to explore.
3. Invite your team
Go to Team → Invite member (team workspaces only). Admins manage members and project access there. See Teams, roles and permissions.
4. Find your way around
| Where | What you'll find |
|---|---|
| Top bar | Workspace switcher, Projects, Team, Integrations, Reports, Billing, Settings; Search (⌘K); Ask agent (⌘J); New (quick-create a test case, bug or run); notifications |
| Project sidebar | Agent; Overview: Focus, Insights; Plan: Requirements, Documents, Test Plans, Releases; Tests: Test Suites, Step Library; Run: Test Runs, API Tests, Pipelines; Track: Bugs, Traceability; Settings |
Every list (requirements, runs, bugs, plans…) has search, filters and sort. The current view is kept in the URL, so you can bookmark it or share it with a teammate. Press / to jump to a list's search box.
A typical workflow
- Add or import requirements (from Jira, Notion, or by hand).
- Let the AI analyse them and generate test cases, then review and approve the cases.
- Group cases in test suites, and optionally in a test plan.
- Start a test run, execute the cases, and file bugs from failures.
- Track readiness per release, and sign it off when it ships.
- Connect CI so automated results flow into Pipelines alongside your manual testing.
How it fits together
| Thing | What it is | How it relates |
|---|---|---|
| Requirement | Something the product must do | Test cases trace back to it; its readiness comes from their results |
| Test case | Steps and expected results that verify a requirement | Lives in a test suite; only approved cases are run and count |
| Test suite | A folder of test cases | Where cases live; plans and runs pick from suites |
| Test plan | What you'll test for a release or feature, and how | A selection of cases plus scope, strategy and risks; runs can start from it |
| Test run | One pass through a set of cases, with a result for each | Manual, or from CI (marked CI); link it to a release so it counts there |
| Release | A set of requirements you plan to ship | Readiness comes only from runs linked to it; a project admin signs it off |
| Pipeline (CI run) | Automated results uploaded from CI | Matched cases are recorded as a CI test run, and on a release's branch they count toward that release |
| API test | HTTP requests with assertions, in their own suites | Runs on demand, on a schedule or from CI |
Readiness means the same thing on every page: see Requirements.
The AI in sqale
- Agent (sidebar, or ⌘J): ask questions and take actions in plain English. Changes to existing data need your approval.
- Test generation on a requirement: writes test cases only from what the requirement states, and asks questions instead of guessing.
- Jira: with Jira connected, comment
/generate-testson a Jira issue to generate test cases for it. - Widget on your staging site: report bugs with a screenshot, and (with Demo to test on) record a walkthrough as a Playwright test.
- MCP: use sqale from Claude Desktop, Cursor and other AI assistants.
All of them use your workspace's AI budget. See The AI agent and MCP.