Let your agent press native controls, wait for the expected state and retain named window captures with accessibility snapshots. The MacOSApp guest executor runs these journeys inside source-installed SwiftVM guests, so your agent can compare what changed without taking over your desktop.
waitFor gigahog.plot · present expect gigahog.breadcrumb.item.0 · “reflow-demo” expect gigahog.mode.flame · present waitFor gigahog.plot · present press “Projects, 33.6 MB” waitFor “alpha, 25.2 MB” · present expect gigahog.breadcrumb.item.1 · present press gigahog.breadcrumb.item.0 · “reflow-demo” waitFor gigahog.breadcrumb.item.1 · absent # Presentation: compare base and injected-change runs Recorded MacOSApp actions, grouped by named capture. Comparison comments describe this presentation, not runtime commands. Select a checkpoint line to inspect its output; preceding lines provide context.
app.run
“launched” · window/launched-1.pngView captureax/launched.json7 of 34 nodes, from another run of this build
timeline.jsonwindow/launched-1.pngapp.run
“zoomed” · window/zoomed-1.pngView captureaction targets · result.jsontimeline.jsonax/002-after.jsonwindow/zoomed-1.pngapp.run
“back” · window/back-1.pngView captureaction targets · result.jsonThe app quit through its own Quit command; cleanup confirmed.
timeline.jsonax/002-after.jsonax/006-after.jsonwindow/zoomed-1.pngwindow/back-1.pngcompare window/zoomed-1.png24×47 px at 244,1176 artifacts
Every step passed. The gauge did not. The accessibility snapshots match, but the needle is drawn mirrored: the disk is more than half full and the needle now leans left. Was the gauge drawing meant to change?
Real MacOSApp window captures and accessibility snapshots from QA runs of Gigahog, a disk-usage app in development, on synthetic folders. Screens from separate runs of one build are shown in sequence; before each press, the run’s other root or zoomed capture stands in for the screen. The needle change was injected on purpose; its crops come from one round of that exercise and the comparison figures from a repeat, outlined here on another run’s capture. The guest executor is used inside source-installed SwiftVM guests; the standalone host provider is a separate development candidate. MacOSApp is not bundled with the public CLI.
A scenario presses buttons, sets values and waits for elements to appear or disappear before it takes a named capture. Each run launches a test copy of your app, quits it through its own Quit command and confirms that every process it started has stopped.
A screen can change while no accessibility text does: a gauge, a chart, an icon or a colour. Keeping the window capture beside its accessibility snapshot lets your agent tell a redrawn control from a renamed or missing one.
In a QA exercise, a deliberately injected change mirrored a disk gauge’s needle. Every step passed and the accessibility snapshots matched; the compared captures differed in one small region. Your agent can bring both captures into review and ask whether the drawing was meant to change.
Bring the evidence into PR reviewThe MacOSApp guest executor has run real app journeys inside SwiftVM through Reflow. It receives Accessibility and Screen Recording permissions from the provisioned guest agent, with a preflight check in each clone. MacOSApp is not bundled with the public CLI. The standalone host provider and its source-catalog recipe are a separate development path; guest execution does not establish that host installation is accepted.
The MacOSApp guest executor is used inside source-installed SwiftVM guests. The standalone host provider remains a separate development candidate; this app.run excerpt describes its interface, not a bundled installation recipe.
---
providers:
app: MacOSApp
---
app.run scenario="{\"version\":1,\"actions\":[{\"do\":\"press\",\"selector\":{\"role\":\"AXButton\",\"label\":\"Blocks\"}},{\"do\":\"waitFor\",\"state\":\"present\",\"selector\":{\"identifier\":\"mode-label\"},\"value\":\"Mode: Blocks\"},{\"do\":\"capture\",\"name\":\"blocks\"}]}"Plugin requirements and private connection settings are configured separately.