A hardened shell execution environment that runs commands within a dedicated, persistent tmux session to ensure safe and isolated terminal interactions.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install claw-shell-hardened
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 claw-shell-hardened using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Claw Shell Hardened is a specialized utility designed for AI agents to interact with system shells through a controlled and persistent tmux environment. By isolating all operations within a single session named claw, it provides a stable workspace where command history and process states are maintained across multiple agent turns.
This skill is a critical component of the Openclaw Skills ecosystem, prioritizing system integrity through a robust security layer. It prevents accidental system damage by enforcing strict guardrails around destructive commands while providing the AI agent with a transparent interface to execute development tasks, system audits, and repository management.
To integrate this capability into your environment, ensure tmux is installed on your host system and add the skill to your agent configuration:
# Ensure tmux is installed
sudo apt-get install tmux
# The skill will automatically handle session creation
# for the 'claw' environment upon first execution.
Claw Shell Hardened utilizes a streamlined data flow focused on session isolation and output capture:
| Component | Type | Description |
|---|---|---|
| Session Name | String | Hardcoded to claw to prevent cross-session interference. |
| Input Payload | String | The raw shell command to be executed. |
| Output Capture | Text Block | The terminal output read directly from the tmux pane buffer. |
| Guardrail Log | Metadata | Records instances where restricted commands (sudo, rm) triggered user prompts. |
Loading
A specialized research skill that enables AI agents to search, summarize, and maintain a persistent log of ArXiv papers.

An automated code analysis and repair tool designed to identify bugs, security vulnerabilities, and performance bottlenecks with optional in-place fixing.

An AI-optimized web search skill that leverages the Tavily API through AIsa's unified gateway to provide clean, relevant data for autonomous agents.

A professional-grade reconnaissance tool for host discovery, service fingerprinting, and CVE matching without external dependencies.

A discovery engine for finding and installing specialized capabilities within the Openclaw Skills ecosystem.

ClawSec Monitor is a transparent HTTP/HTTPS proxy designed to inspect, log, and secure all real-time traffic generated by AI agents.








































