blackbox

Coordinate multiple AI models to generate and judge code implementations.

2|Updated Apr 25, 2026
One-click install
npx skills add https://github.com/AlexiosBluffMara/mercury --skill blackbox-alexiosbluffmara
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/AlexiosBluffMara/mercury/tree/main/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/AlexiosBluffMara/mercury --skill blackbox-alexiosbluffmara

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates complex coding tasks to a multi-model AI CLI, eliminating manual coding bottlenecks and accelerating prototyping.

Core Features & Use Cases

  • Multi-model evaluation: tasks are dispatched to Claude, Codex, Gemini, and Blackbox Pro with a built-in judge that selects the best implementation.
  • Interactive and non-interactive sessions with PTY support, including checkpointing and resume capabilities.
  • Background mode and parallel task execution for long-running or concurrent tasks with progress monitoring.
  • Vision model switching and flexible model configuration to adapt to different tasks.

Quick Start

Prompt Blackbox to build a small REST API, for example 'Build a REST API for todos with SQLite'.

Frequently Asked Questions about blackbox

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

FAQPage Schema
How do I use a multi-model AI CLI for automated code generation?

Automated code generation with a multi-model AI CLI dispatches tasks to models like Claude, Codex, and Gemini, using a built-in judge to select the best implementation for your software projects.

What do I need to configure before delegating coding tasks to a multi-model AI?

Delegating coding tasks requires the Blackbox CLI installed and configured with a valid API key to enable interactive PTY sessions, background execution, and model switching.

Can I run long-running coding tasks in the background with an AI coding agent?

Background mode supports parallel task execution for long-running or concurrent coding tasks, providing progress monitoring so you can track automated prototyping and code generation.

Does multi-model AI code generation work for debugging and code review?

Multi-model AI code generation applies to automated debugging and code review by coordinating multiple models to generate, evaluate, and judge code implementations across projects.

How does an AI coding agent handle interactive sessions and checkpoints?

Interactive PTY sessions support checkpointing and resume capabilities, allowing you to pause long-running multi-model code generation tasks and continue them later without losing progress.

What is the best way to prototype a REST API using an AI coding agent?

Prototyping a REST API with an AI coding agent involves prompting the CLI with your project requirements, dispatching the task to multiple models, and letting the built-in judge select the best implementation.