blackbox

Orchestrate multiple AI models for coding tasks via the Blackbox CLI.

Updated Apr 19, 2026
One-click install
npx skills add https://github.com/gqf2008/hermez-ai --skill blackbox-gqf2008
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/gqf2008/hermez-ai/tree/main/skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/gqf2008/hermez-ai --skill blackbox-gqf2008

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Blackbox enables teams to delegate coding tasks to a multisystem AI agent, reducing turnaround time and improving solution quality by leveraging multiple models and an automated judge.

Core Features & Use Cases

  • Multi-model execution and judging: tasks are run across several models and evaluated to pick the best result.
  • Interactive and batch workflows: supports interactive sessions, non-interactive one-shot tasks, and background processing with checkpointing.
  • Collaboration and review: facilitates parallel work, PR-like reviews, and rapid iteration across codebases.

Quick Start

Run a sample coding task with Blackbox CLI by providing a prompt and your API key to see multi-model results.

Frequently Asked Questions about blackbox

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

FAQPage Schema
How do I automate coding tasks using multiple AI models?

You can automate coding tasks across multiple AI models by orchestrating them through a single CLI, which executes your prompt across different models and evaluates the outputs to select the best result.

What is multi-model execution and judging in an AI coding agent?

Multi-model execution and judging is a process where a coding task is run across several AI models simultaneously and an automated judge evaluates the generated solutions to pick the highest quality code.

Do I need Node.js to run multi-model AI coding automation?

Yes, you need a Node.js 20+ environment to run this coding automation, along with the Blackbox CLI installed and a valid API key to access the multiple AI models.

Can I run background coding tasks and resume them later with checkpointing?

Yes, you can run background coding tasks with checkpointing enabled, allowing you to save execution state, resume non-interactive one-shot tasks, and switch models during interactive sessions.

What is the best way to delegate codebase updates for parallel review?

The best way to delegate codebase updates for parallel review is using a multi-model AI agent that facilitates parallel work, generates PR-like reviews, and enables rapid iteration across your codebases.

Why use an automated judge for AI-generated code instead of manual review?

Using an automated judge for AI-generated code reduces turnaround time and improves solution quality by systematically evaluating results from multiple models, which is more efficient than manual review.