Verify macOS UI focus and confirm actions with screenshots.
npx skills add https://github.com/Kira-Pgr/axiclick --skill axiclick
Please help me install this Agent Skill. Skill: axiclick Source: https://github.com/Kira-Pgr/axiclick/tree/main/skill Command: npx skills add https://github.com/Kira-Pgr/axiclick --skill axiclick
axiclick prevents unreliable desktop automation by forcing agents to verify the correct focused window and confirm UI state after each action, avoiding blind clicks and silent failures.
active checks before typing/clicking and provides a protocol for wrong-window retries.som to detect and label UI elements, then som-click @<id> to click precisely without fragile coordinates.screenshot verification after clicks, typing, navigation, scrolling, and app switching; includes probe for ambiguous labels.focus <app>, accessibility tree reads via snapshot, and session setup via som-start.Run axiclick’s standard workflow to bring an app to the front, verify focus, run SoM to label UI elements, click the target by ID, then take a verification screenshot.
High-intent search queries and answers about installing and using this skill.
To prevent blind clicks in macOS desktop automation, verify window focus and use vision-based UI element selection before clicking. Capturing a screenshot after each action confirms the result, avoiding silent failures across multi-step agent workflows.
Vision-based UI element selection uses a model to detect and label on-screen elements, allowing agents to click by ID rather than fragile coordinates. This approach reads the accessibility tree to navigate interfaces reliably.
To click typed UI elements without coordinates, run a vision model to label interface elements, then target the specific element by its assigned ID. This requires checking that the correct window is active before interaction.
Yes, LLM agent desktop automation works with macOS focus verification by enforcing active window checks before typing or clicking. If the wrong window is focused, the workflow triggers a recovery protocol to retry safely.
To verify app control results in multi-step automation sequences, take a mandatory screenshot after every click, typing action, navigation, or app switch. For ambiguous UI labels, use a probe to clarify the target before proceeding.
UI automation fails when switching apps if it lacks focus verification and post-action visual confirmation. Ensuring the target app is brought to the front and validating the interface state after switching prevents these silent failures.
PlaidmustacheAutomate background macOS desktop interactions via the computer_use tool.
buiminhduc23011Diagnose deployment failures and restore code to a stable state.
AllBeingsFutureLocate desktop UI elements via UIAutomation with vision fallback and return coordinates, size, and attributes.
RheasilviaAutomate native desktop application interactions via accessibility trees and input events.