Omarchy 4.0.3 × Midscene

Run 36207887440 · 93.0% passed · GitHub Actions · Report history

Each image is the original page screenshot used by that node. Click a case or image to open the exact Midscene step and Agent replay.

Doubao Say

13/14 cases passed

Open native Midscene report 1 → Open native Midscene report 2 → Open native Midscene report 3 → Open native Midscene report 4 →

Result	Case	Duration	Node screenshot	AI response / error
✅ Passed	Cancel sign-in and retry	3m48s		AI: The screenshot shows a Doubao Say page with a microphone icon in the center and a list of microphone options below. The selected option appears to be "Synthetic microphone one" (合成麦克风一), and the page is in the Microphone step as indicated by the step indicator at the top.
✅ Passed	Complete Doubao Say onboarding on a real desktop	8m40s		AI: The screenshot shows a settings page with a "Setup" section. There is a disabled control (grayed out) that displays the text "Setup completed by the synthetic E2E fixture." This matches the statement exactly.
✅ Passed	Recover from an endpoint failure	2m01s		AI: The screenshot shows a page with a "Synthetic" section. There is a green checkmark icon next to "Synthetic" and a message indicating "All systems operational". Below that, there is a table with columns for "Endpoint", "Status", and "Response time". The table lists several endpoints, all showing a green "Operational" status. There is no visible error message or indication of an "unavailable" error. The successful endpoint results (status and response times) are clearly visible.
✅ Passed	Changing microphone requires another check	6m27s		AI: The screenshot shows a setup wizard interface with a "Finish setup" button at the bottom. The button appears to be visually distinct, with a blue background and white text, indicating it is in an enabled state rather than grayed out or disabled.
✅ Passed	Capture and retain a custom trigger shortcut	3m01s		AI: The screenshot shows a settings page with a "Trigger" section. The active trigger is displayed as "⌃ F8" (Control + F8) with a "Selected" badge, and the available triggers list below shows "⌃ F8" as well. The page appears to be the Trigger key configuration page, and the active/selected trigger is indeed ⌃ F8.
✅ Passed	Toggle polishing without losing configuration	1m50s		AI: The screenshot shows a settings panel with a blue toggle switch labeled "Voice polishing" that appears to be in the ON position (blue color). Below it, there is a "Model" text field that visibly reads "synthetic-model".
✅ Passed	Complete and cancel a voice test safely	6m08s		AI: The screenshot shows a settings page with a "Voice test" section. The "Start voice test" button is visible and appears enabled (not grayed out). The setup control displays "Voice test passed" text. There is no indication of anything being pasted or sent in the visible content. The statement appears to be true based on the visible elements.
✅ Passed	Switch to Volcengine and validate credentials	5m19s		AI: The screenshot shows a microphone setup page with a title "Let's make sure we can hear you." and a microphone icon. The page appears to be part of a voice assistant or audio setup flow, with a "Continue" button at the bottom. The statement matches the visible content on the screen.
❌ Failed	Switch to Deepgram English recognition	2m00s		Error: Node "aiAssert" failed: Node "aiAssert" failed: Assertion failed: The page is Recognition, explains Deepgram Nova-3 English dictation, shows an empty API Key field, and hides Open Doubao sign-in. Reason: The screenshot shows a page with the title "Recognition" and a description mentioning "Deepgram Nova-3 English dictation". The API Key field appears to be empty. There is no visible "Open Doubao sign-in" element on the page; instead, the interface shows a "Sign in with Doubao" button, which is not hidden but present. Therefore, the statement that "Open Doubao sign-in" is hidden is false.
✅ Passed	Deliver one synthetic dictation to the focused target	1m39s		AI: The screenshot shows a page with a field labeled "Synthetic dictation" containing the value "Synthetic dictation delivered exactly once". Below it, there is a status section showing "Delivered once" with a count of "1 delivery". The information matches the statement exactly.
✅ Passed	Preserve signed-in account across backward navigation	3m28s		AI: The screenshot shows a microphone setup step in what appears to be a voice assistant or speech recognition configuration flow. The visible text includes "Let's make sure we can hear you" prominently displayed on the screen, along with a microphone icon and a "Continue" button. This matches the statement about the Microphone step showing that exact phrase.
✅ Passed	Require an enabled trigger and retain the replacement	2m20s		AI: The screenshot shows a configuration panel for a trigger step. The "Trigger" key step section displays "Active trigger F8" with a dropdown menu. The dropdown appears to be open or showing its current selection, and "F8" is visible as the selected option in the dropdown list. The text "Active trigger F8" is clearly displayed, and the dropdown shows "F8" as the selected value.
✅ Passed	Microphone check gates setup completion	3m41s		AI: Based on the screenshot, I can see a setup wizard interface with a "Finish setup" button at the bottom right. The button appears to be styled with a blue background and white text, which typically indicates an enabled state. There is no visual indication (such as grayed-out styling or a disabled appearance) that would suggest the button is disabled.
✅ Passed	Cancel synthetic dictation without delivering text	3m21s		AI: The screenshot shows a user interface with a section labeled "Synthetic Sentences" and a status area. There is a list of sentences with checkboxes, and a status indicator showing "1 delivered" with a green checkmark. The interface appears to be in a state where a new attempt has been made, and the status reports exactly one delivery. The sentence list shows one item that appears to be selected or delivered, matching the status of "1 delivered". The overall layout suggests a successful operation with a single delivered synthetic sentence.

Omarchy visual checks

1/1 cases passed

Open native Midscene report →

Result	Case	Duration	Node screenshot	AI response / error
✅ Passed	System menu text and focused row are visible	1m00s		AI: The screenshot shows a desktop environment with a vertical bar on the left edge. The bar is tall, extending from the top to the bottom of the screen, and is positioned along the left side. It contains icons and appears to be a dock or launcher. There is no horizontal bar visible at the top or bottom of the screen.