Shell Execution

Execute shell commands scoped to the active workspace with background support.

577|62|Updated May 15, 2026

Search Tags:#automation #audit #safety #workspace #execution #command #shell

Ready to run

npx skills add https://github.com/agentic-in/elephant-agent --skill shell-execution

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: Shell Execution
Source: https://github.com/agentic-in/elephant-agent/tree/main/packages/skills/builtin_packages/runtime/shell-execution
Command: npx skills add https://github.com/agentic-in/elephant-agent --skill shell-execution

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Keeps explicit shell work legible, bounded, and attached to the current workspace thread.

Core Features & Use Cases

Scope commands to the active workspace unless the user clearly asks for something broader.
Explain risky commands and surface concrete results instead of vague summaries.
When a command should keep running, support background execution with explicit follow-up inspection.

Quick Start

Ask me to run a specific shell command in the current workspace and I will execute it with safety and context.

Frequently Asked Questions about Shell Execution

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I execute shell commands safely within an active project workspace?▼

Shell execution tools scope commands to the active workspace thread, ensuring automated operations remain legible, bounded, and auditable throughout iterative project steps.

What is the best way to audit risky shell commands during automation?▼

Auditing risky shell commands involves explaining the command's impact before execution and surfacing concrete results, keeping explicit shell work clear and traceable within the workspace.

Can I run long-running shell commands in the background and inspect them later?▼

Background execution supports long-running shell commands by allowing them to keep running while providing explicit follow-up inspection mechanisms to check status and results later.

Does shell execution work with iterative tasks across a single workspace?▼

Shell execution is designed for iterative tasks within a single workspace, attaching commands to the current project thread to maintain context and safety across multiple steps.

When should I scope shell command execution to the active workspace?▼

Scope shell command execution to the active workspace by default unless explicitly asked for broader access, preventing unintended side effects and maintaining bounded, auditable operations.