shell

Executes user-provided text after /shell as a literal terminal command.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/shoshoavi/agentic_worflows --skill shell-shoshoavi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: shell
Source: https://github.com/shoshoavi/agentic_worflows/tree/main/cursor/skills-cursor/shell
Command: npx skills add https://github.com/shoshoavi/agentic_worflows --skill shell-shoshoavi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you explicitly want a command run verbatim in the terminal, AI assistants often rewrite, explain, or second-guess the command first. This Skill guarantees that text following a /shell invocation is executed exactly as typed, without modification or unnecessary repository inspection. ## Core Features & Use Cases - Literal Command Execution: Treats all text after /shell as the exact shell command and runs it immediately with the terminal tool. - No Rewriting or Inspection: Skips command rephrasing, explanation, and repository scanning unless the command itself requires repository context. - Concise Reporting: Reports exit status and important stdout or stderr after execution. - Use Case: Type /shell ls -la src/ to list directory contents instantly, or /shell git status to check your working tree without the AI altering the command. ## Quick Start Invoke /shell followed by the exact command you want run, such as asking the assistant to run /shell npm test to execute your test suite verbatim.

Frequently Asked Questions about shell

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

FAQPage Schema
How do I run an exact shell command through an AI assistant?▼

Invoke /shell followed by the exact command text. The assistant treats everything after the invocation as the literal command and executes it immediately with the terminal tool, without rewriting or explaining it first.

What happens if I invoke /shell with no command?▼

If /shell is invoked without any following text, the assistant asks which command you want to run instead of guessing or executing anything on its own.

Will the assistant modify my command before running it?▼

No. The Skill explicitly forbids rewriting, improving, or explaining the command before execution. It also skips repository inspection unless the command itself requires repository context.

What output do I get after the command runs?▼

After execution, the assistant briefly reports the exit status along with any important stdout or stderr produced by the command, keeping the response concise.

When should I not use the /shell invocation?▼

Avoid /shell when you want the assistant to interpret a goal and choose commands itself, or when the task needs multi-step reasoning. It is designed only for explicit, verbatim command execution.