omniCell

Execute system commands and Nushell scripts, returning structured JSON results.

15|4|Updated Dec 24, 2025
One-click install
npx skills add https://github.com/tao3k/omni-dev-fusion --skill omnicell
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omniCell
Source: https://github.com/tao3k/omni-dev-fusion/tree/main/assets/skills/omniCell
Command: npx skills add https://github.com/tao3k/omni-dev-fusion --skill omnicell

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

OS interactions often produce unstructured text that is hard to consume programmatically. OmniCell converts commands and system data into clean JSON objects for reliable downstream processing.

Core Features & Use Cases

  • Structured JSON output for system commands and Nushell scripts.
  • Safe, intent-driven execution with observe (read) and mutate (write) modes.
  • Use cases include querying system state, executing commands, and integrating with automation agents.

Quick Start

Run a Nushell command through the execute tool to obtain a JSON result.

Frequently Asked Questions about omniCell

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

FAQPage Schema
How do I get structured JSON output from system commands?

Structured JSON output from system commands is obtained by executing commands through an intent-driven processor that returns consistent objects containing status, data, and action fields for reliable downstream consumption.

How do I execute Nushell scripts and return JSON results?

Nushell scripts are executed by passing them to the execution tool, which processes the system commands and returns the results as clean JSON objects instead of unstructured text.

What is the best way to query system state for automation agents?

Querying system state for automation agents is best handled by classifying intents into observe or mutate modes, executing the system interaction, and returning structured JSON results for programmatic use.

Why does parsing OS interaction output programmatically fail?

Parsing OS interaction output fails because system commands typically produce unstructured text that lacks a consistent schema, making it difficult to extract reliable data for downstream processing without converting it to JSON.

Does intent classification affect command execution safety?

Intent classification affects safety by separating observe or read modes from mutate or write modes, ensuring that system commands are executed securely and return structured JSON results without unintended modifications.