openclaw-sandbox

Community

Securely isolate tool execution.

Authorollieb89
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill provides robust Docker-based sandboxing for tool execution, ensuring that potentially risky operations are isolated from the host system, enhancing security and stability.

Core Features & Use Cases

  • Configurable Isolation: Define the scope and mode of sandboxing (session, agent, shared) for different execution contexts.
  • Workspace Management: Control how agents access their workspace within the sandbox (none, read-only, read-write).
  • Customization: Mount host directories into the sandbox and configure network access and environment variables.
  • Use Case: When running untrusted code or tools that require specific dependencies, configure the sandbox to use a read-only workspace and no network access to prevent unintended side effects.

Quick Start

Configure the sandbox to use 'non-main' mode with 'session' scope and 'rw' workspace access.

Dependency Matrix

Required Modules

None required

Components

references

💻 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: openclaw-sandbox
Download link: https://github.com/ollieb89/openrepo/archive/main.zip#openclaw-sandbox

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.