human-shell

Execute shell commands with output buffering and safety validation.

Updated May 6, 2026

Search Tags:#automation #context #security #execution #terminal #shell

Ready to run

npx skills add https://github.com/dbowm91/codegg --skill human-shell

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: human-shell
Source: https://github.com/dbowm91/codegg/tree/main/.opencode/skills/human-shell
Command: npx skills add https://github.com/dbowm91/codegg --skill human-shell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of executing shell commands within an AI agent environment while maintaining strict control over context window usage and system security.

Core Features & Use Cases

Context Management: Run commands without polluting the model's conversation history unless explicitly promoted.
Safety Guardrails: Automatically blocks dangerous commands like system shutdowns or recursive deletions and provides confirmation for sensitive operations.
Output Handling: Captures and stores command output in a bounded, searchable buffer, allowing for structured failure analysis and selective inclusion.

Quick Start

Execute a shell command by typing an exclamation mark followed by your command in the prompt to run it ephemerally without adding the output to the conversation context.

Frequently Asked Questions about human-shell

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

FAQPage Schema

How do I execute shell commands without polluting the model context window?▼

You can run shell commands without polluting the model context by executing them in a controlled environment that uses explicit promotion policies. Commands run ephemerally, storing output in a bounded buffer unless explicitly added to the conversation.

What is the best way to prevent dangerous terminal commands from running in an AI agent?▼

Preventing dangerous terminal commands requires strict security filtering and safety validation during execution. The system automatically blocks operations like system shutdowns or recursive deletions and requires confirmation for sensitive modifications.

Can I execute terminal commands ephemerally without saving the output to conversation history?▼

Yes, you can execute terminal commands ephemerally by prefixing your command with an exclamation mark. This runs the shell command without adding the output to the conversation context, keeping your history clean.

How does output buffering work for terminal-based agent interactions?▼

Output buffering for terminal-based agent interactions works by capturing and storing command results in a bounded, searchable buffer. This allows for structured failure analysis and selective inclusion without causing context overflow.

Does this shell execution approach work for analyzing failed commands without context overflow?▼

Yes, this shell execution approach works for analyzing failed commands by capturing output in bounded storage. This prevents context overflow while allowing you to search the buffer and selectively promote relevant failure details to the model.

When should I use a controlled shell execution environment instead of direct terminal access?▼

You should use a controlled shell execution environment instead of direct terminal access when managing AI agent interactions. It provides necessary safety guardrails, bounded storage, and context management to prevent unauthorized system modifications.