A sophisticated automation suite that executes, chains, and manages AI applications on transweb.cn through browser-based agent workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-web-cpu
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agent-web-cpu using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is agent-web-cpu?
agent-web-cpu is a high-performance skill designed to bridge the gap between user requirements and the automated execution of web-based AI tools. It functions as an intelligent wrapper for transweb.cn, allowing users to trigger complex AI workflows through simple natural language commands. By leveraging Openclaw Skills for deep browser integration, the agent can autonomously navigate web interfaces, input data using simulated keyboard events, and extract structured results into clean Markdown format. This skill is essential for developers looking to automate repetitive content generation tasks or build complex multi-stage AI pipelines without manual browser interaction.
agent-web-cpu Use Cases
- Creating automated content pipelines that generate an outline first and then expand into a full blog post.
- Registering and managing a library of custom AI tools by their unique transweb.cn IDs.
- Batch processing long-form text for polishing, rewriting, or summarization through browser-based AI agents.
- Extracting structured Markdown reports from web-based AI generation tools for documentation purposes.
How agent-web-cpu Works
- The skill analyzes user input to decide between single-app execution, multi-app pipelines, or registry management modes.
- It performs a status check on the browser tool to ensure that navigation and JavaScript evaluation are enabled within the Openclaw environment.
- The agent searches the local apps.json registry using a weighted matching algorithm to find the most relevant AI application.
- If the input exceeds 500 characters, it utilizes advanced simulated typing to bypass frontend framework restrictions; otherwise, it uses URL parameter injection.
- It monitors the browser DOM for specific result containers, waiting for the AI generation to complete within a 30-second timeout window.
- Finally, it uses a custom JavaScript extraction script to parse the output and save it as a timestamped Markdown file in the local downloads directory.
agent-web-cpu Setup
To deploy this skill, ensure your OpenClaw configuration allows for full browser control. Essential steps include:
- Enable the browser tool in your configuration settings.
- Set
browser.enabledandbrowser.evaluateEnabledtotrueto allow for result extraction. - Verify the installation by running the status check command:
browser action: status
- Ensure the
apps.jsonfile is present in the skill directory to track registered AI applications.
agent-web-cpu Data Schema & Taxonomy
The skill manages its application registry through a JSON-based schema in apps.json. The data is structured as follows:
| Property | Type | Description |
|---|---|---|
| id | String | Unique 32-character hex identifier for the transweb app. |
| name | String | Human-readable name used for command matching. |
| description | String | Detailed metadata describing the application's purpose. |
| keywords | Array | List of terms used to increase matching accuracy during user requests. |
| createdAt | String | ISO 8601 timestamp indicating when the app was registered. |
agent-web-cpu Advanced Features
- Multi-stage Pipeline Execution: Seamlessly pipes the output of one AI application as the input for the next in a sequence.
- Intelligent Typing Simulation: Uses simulated hardware events to ensure compatibility with modern reactive frameworks like Vue and React.
- Automatic App Discovery: If a requested app is missing, the skill can navigate to the ID-specific URL to scrape metadata and register it automatically.
- Structured Markdown Extraction: Specifically parses H1, H2, and pre tags to preserve document hierarchy during extraction.
- Flexible Browser Profiles: Supports the standard openclaw sandbox profile for security and the user profile for sessions requiring authentication.
SKILL.md
Loading
Related Openclaw Skills
OpenRelay Agent-to-Agent Marketplace
A comprehensive skill for AI agents to register, publish services, and conduct autonomous transactions within a decentralized marketplace.
Snail
Snail is an AI agent dating protocol that prioritizes cautious matching and slow-paced connection building.
Duck Dating for AI Agents
A specialized skill enabling AI agents to form steady, reliable connections on inbed.ai through a duck-philosophy matching algorithm.
Goose Dating
Goose Dating is a specialized social protocol for AI agents designed to foster deep loyalty, commitment, and protective relationships.
Article Formatter - Professional Self-Media Content Styler
A specialized tool for formatting self-media articles into highly readable, visually appealing, and mobile-optimized content.
lsp-assist
A precise Language Server Protocol integration for Openclaw Skills that enables deep code intelligence and navigation.
Featured*
