A powerful Python-based controller for multi-account browser management via HTTP API.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install account-switcher
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install account-switcher using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Account Switcher is a specialized skill designed to interface with multi-account matrix management tools. It enables AI agents and developers to programmatically control browser instances, allowing for the automated starting, stopping, and status querying of specific account profiles through a local HTTP API.
By leveraging Openclaw Skills, this tool bridges the gap between high-level AI commands and local browser automation. It features intelligent matching logic, supporting both precise and fuzzy name matching to ensure reliable execution even with complex account naming conventions.
To get started with this skill in your Openclaw Skills workflow, ensure you meet the prerequisites:
Environment Setup: Install Python 3.8+ and the required requests library:
pip install requests
Matrix Tool Configuration: Install the Multi-Account Matrix Management Tool. Open the software and navigate to Settings -> Local Server, then check Enable Local Server on port 1008.
Skill Verification: Check your Python version to ensure compatibility:
python --version
The skill manages data through a local configuration and a temporary result tracker:
| File | Description |
|---|---|
| config.yaml | Contains the local API endpoint (default: http://localhost:1008) and timeout settings. |
| matrix_manager.py | The core logic handler for all Openclaw Skills account operations. |
| last_result.txt | A UTF-8 encoded text file storing the output of the most recent command execution. |
Note on Indexing: The internal API uses 0-based indexing, but this skill maps operations to a user-friendly 1-based index (e.g., Index 1 in the command corresponds to API Index 0).
Loading
An autonomous iterative framework for AI agents to solve multi-step problems through a structured Plan-Act-Evaluate loop.

An automated tool for auditing, identifying, and fixing web accessibility violations to ensure inclusive user experiences.

A high-performance rendering framework for optimizing 2D Canvas applications handling over 5,000 elements.

A specialized developer productivity toolkit designed to automate project scaffolding, enforce code standards, and streamline complex development lifecycles.

A synthesis-driven AI agent skill for generating professional social media posters, speaker announcements, and partnership content.

A comprehensive digital companion for exploring Hong Kong's culture, landmarks, and essential travel logistics.








































