A secure file management utility designed to perform read, write, and search operations within a strictly controlled sandbox environment.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install workspace-files
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 workspace-files 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 Workspace Files Skill?
The Workspace Files skill is a fundamental component for any developer using Openclaw Skills, providing a secure abstraction layer for file system interactions. It ensures that all file operations—including directory listing, reading, writing, and searching—are confined to the ~/.openclaw/workspace directory, preventing unauthorized access to the host system.
By utilizing this skill, AI agents can maintain a clean and organized project structure while adhering to strict security protocols. This sandboxed approach makes it an essential tool for automated coding tasks, documentation management, and log processing without risking system integrity.
Workspace Files Skill Use Cases
- Automating the creation of project documentation and metadata files within a secure environment.
- Searching for specific configuration files or skill definitions across a complex directory structure.
- Reading logs or output data generated by other Openclaw Skills for further processing.
- Safely initializing new workspace directories and subfolders for organized agent workflows.
How Workspace Files Skill Works
- The AI agent initiates a command via the workspace-files.sh script, providing a relative path or search pattern.
- The skill validates the requested path to ensure it resides strictly within the ~/.openclaw/workspace sandbox.
- The system executes the requested operation (list, read, write, or search) using the provided parameters.
- The result, such as file content or a list of directory items, is returned to the agent in a structured format.
- For write operations, the skill ensures the persistence of text data for future retrieval or cross-agent communication.
Workspace Files Skill Setup
To integrate the Workspace Files skill into your environment, ensure your workspace directory is correctly initialized. Use the following commands to test the installation:
# Verify the scripts directory exists and list the current workspace
./scripts/workspace-files.sh list-dir "."
# Create a test file to confirm write permissions within the sandbox
./scripts/workspace-files.sh write-file "07_OUTPUTS/init.txt" "Initialization complete"
Workspace Files Skill Data Schema & Taxonomy
The skill operates on a hierarchical directory structure rooted at /home/cmart/.openclaw/workspace. It categorizes data through the following organization:
| Directory/Component | Description |
|---|---|
01_AGENTS |
Storage for agent-specific configurations and registries. |
07_OUTPUTS |
Default destination for generated text files and test results. |
scripts/ |
Contains the workspace-files.sh execution logic. |
Relative Paths |
All file operations use paths relative to the workspace root for portability. |
Workspace Files Skill Advanced Features
- Sandboxed Security: Strict enforcement of path boundaries to prevent directory traversal attacks.
- Pattern-Based Searching: Efficiently locate files using partial string matches across the entire workspace.
- Multi-Agent Collaboration: Enables different agents to share data by reading and writing to common directories within Openclaw Skills.
- Directory Inspection: Quick recursively listing of directories to map out complex project structures for the AI agent.
SKILL.md
Loading
Related Openclaw Skills
Pricing Strategy
A comprehensive framework for designing, optimizing, and communicating SaaS pricing models to maximize revenue and customer value.
Football Player DNA Similarity Search
An AI-driven scouting skill that uses vector similarity to find football players with matching playing styles and physical attributes at better market values.
Nano Banana Pro
A professional-grade image generation and editing tool leveraging Gemini 3 Pro for AI-driven visual workflows.
Zoom Phone MCP
A read-only Model Context Protocol (MCP) server that connects AI agents to the Zoom Phone REST API.
x-osv: Open Source Vulnerability Scanner
A powerful CLI extension for querying the Google OSV database and scanning local projects for package vulnerabilities.
Session Cleaner
Automated maintenance utility to archive stale transcript files while preserving active main and group chat sessions.
Featured*
