blackbox

Delegate coding tasks to a multi-model AI agent with an integrated judge.

3|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/ever-oli/io --skill blackbox-ever-oli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/ever-oli/io/tree/main/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/ever-oli/io --skill blackbox-ever-oli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill delegates coding tasks to Blackbox AI CLI, enabling multi-model evaluation and selection of high-quality implementations to accelerate software development.

Core Features & Use Cases

  • Delegates coding tasks to multiple models via the Blackbox CLI and uses an integrated judge to pick the best result.
  • Supports interactive sessions, background tasks, parallel work, checkpoints, and resume capabilities for long-running tasks.
  • Useful in code review, rapid prototyping, and solution verification across model providers and configurations.

Quick Start

Start a Blackbox task by prompting the CLI to generate an implementation for a feature and compare results across models.

Frequently Asked Questions about blackbox

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

FAQPage Schema
How do I compare multiple AI model outputs for code generation?

You can delegate coding tasks to a multi-model AI judge that runs tasks through several models and a built-in judge to generate and compare multiple implementations, selecting the best result.

What is the best way to manage long-running AI coding tasks in the background?

Managing long-running AI coding tasks is done via a CLI-based agent that supports background tasks, parallel work, checkpoints, and resume capabilities to ensure continuous session management.

Can I use a CLI to orchestrate multi-provider AI agents for software development?

Yes, you can use a CLI to orchestrate multi-provider AI agents for software development, satisfying requirements for multi-provider orchestration, session management, and checkpointed workflows with proper safeguards and configuration.

Does multi-model AI code generation work for rapid prototyping and solution verification?

Multi-model AI code generation works effectively for rapid prototyping and solution verification by generating and comparing multiple implementations across different model providers and configurations.

How to resume interrupted batch coding tasks across multiple models?

To resume interrupted batch coding tasks across multiple models, use an AI agent with checkpointed workflows that allow you to save progress and resume long-running tasks during interactive sessions.

When should I not use a multi-model AI judge for code review?

You should not use a multi-model AI judge for code review if your environment lacks proper safeguards and configuration for CLI-based control, or if your task does not require comparing implementations across multiple providers.