axiclick

Community

Click safely with vision-based IDs—no guessing.

AuthorKira-Pgr
Version1.0.0
Installs0

System Documentation

What problem does it solve?

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.

Core Features & Use Cases

  • Focus verification and recovery: Enforces active checks before typing/clicking and provides a protocol for wrong-window retries.
  • Vision-based UI interaction (SoM): Uses som to detect and label UI elements, then som-click @<id> to click precisely without fragile coordinates.
  • Post-action visual confirmation: Requires screenshot verification after clicks, typing, navigation, scrolling, and app switching; includes probe for ambiguous labels.
  • Practical app control loop: Supports launching/unminimizing via focus <app>, accessibility tree reads via snapshot, and session setup via som-start.

Quick 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.

Dependency Matrix

Required Modules

None required

Components

assets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: axiclick
Download link: https://github.com/Kira-Pgr/axiclick/archive/main.zip#axiclick

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.