Skip to main content

Editor toolbar

The plugin puts a toolbar of its own above each spec file, so the things you do to a spec are one click away rather than somewhere in a menu. It appears only on the file extensions the plugin handles, and you can turn it off with Show the editor toolbar above spec files.

The left side holds the actions for the spec file itself, in this order: the AI agent button, Run, the status lozenges, the folding buttons, and the report timestamp. The right side holds the toggles for the panels that open inside the editor. In split view, the right side gains two more controls for the Execution Report pane — scroll sync and link highlight — which are hidden when not in split view mode.

<The toolbar above an open spec file, with the status lozenges and report timestamp on the left and the panel toggles on the right.>

Launch an AI agent​

An opt-in button opens a terminal session with an AI coding agent, either Claude Code or GitHub Copilot CLI. The plugin remembers that session against the spec file you are editing, so a later click brings back this spec's own session rather than starting another one. Refer to AI agent integration.

<The agent button on the toolbar with its dropdown open, showing the choice of agent, beside a terminal tab running a session for the open spec file.>

Run this spec file​

A Run button runs the tests of the whole spec file, wherever the caret happens to be, so you do not scroll up to the Feature: line first. The plugin finds the right class to run whichever SpecBinder generation mode you use. Until a generated test class exists, the button is disabled.

It ships with a default shortcut, registered under its own SpecBinder group in the Keymap and chosen not to collide with anything built into IntelliJ.

<The Run button on the toolbar, with the caret part way down a spec file, and the run window below showing the scenarios of that file.>

Status lozenges​

Coloured lozenges count the scenarios of the file by result: passed, failed, not executed.

Click a lozenge to move the caret to the next scenario with that result, and the editor scrolls to it. After the last one it returns to the first.

The counts come from one of two places. With execution results enabled they come from SpecBinder's report; otherwise they come from the IDE's own record of the last test run.

<The toolbar lozenges on a file that has run — a passed count and a failed count side by side, with no lozenge for a result that has no scenarios.>

Expand and collapse everything​

Two buttons collapse and expand every Scenario and Background in the file at once. Collapsing leaves just the title lines, which turns a long spec into a readable outline you can scan and then open only where you need to.

<A spec file with every Scenario folded to its title line, and the two folding buttons on the toolbar above it.>

Report status​

Beside the lozenges, the toolbar shows when the loaded execution report was last updated.

When a run produces a newer report, a Reload link appears beside the timestamp. Click it to move to the new results, or carry on reading the ones you have. To skip that click, turn on auto-reload — a toggle on the Execution Report pane — and each new report loads by itself.

<The report timestamp on the toolbar with a Reload link beside it, after a newer report has arrived.>

Inline panels under steps​

The plugin can open a panel inside the editor, directly under a step: the error panel on a failed step, the values a step published through JUnit's TestReporter while it ran, and the preview of doc string content for images, SVGs and HTML. Three split buttons on the right of the toolbar control the three kinds.

The main half of each button opens or closes every panel of that kind in the file, and stays pressed while any of them are open. The dropdown half carries that family's appearance settings, starting with the colour of the panel's left bar.

<The dropdown of one panel toggle open, showing its appearance settings.>