blackbox

Execute coding tasks through multiple LLMs and select the best result.

2|Updated May 22, 2026
One-click install
npx skills add https://github.com/519lab/thoth-agent --skill blackbox-519lab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/519lab/thoth-agent/tree/main/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/519lab/thoth-agent --skill blackbox-519lab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @blackboxai/cli, and includes scripts (resource) components.

What problem does it solve?

This Skill automates coding tasks by delegating them to the Blackbox AI CLI agent, reducing manual coding time and improving accuracy with multi-model collaboration.

Core Features & Use Cases

  • Multi-Model Coding Agent: Runs tasks through multiple LLMs (Claude, Codex, Gemini, Blackbox Pro) and selects the best result.
  • Interactive Sessions: Supports interactive coding sessions for collaborative development.
  • Background Mode: Handles long tasks in the background, allowing for progress monitoring.
  • Checkpoint & Resume: Pauses and resumes tasks for efficient workflow management.
  • Vision Support: Automatically detects images in input and can switch to multimodal analysis.
  • Use Case: For a developer working on a complex codebase, Blackbox can be used to generate code snippets, refactor modules, or implement new features.

Quick Start

Delegate a coding task to Blackbox AI by running the command: blackbox --prompt 'Implement JWT authentication with refresh tokens in the Express API'.

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?

Automating coding tasks with multiple AI models involves executing prompts through various LLMs like Claude, Codex, and Gemini, then automatically selecting the best generated result. This multi-model approach improves code accuracy and reduces manual effort.

Can I run long code generation processes in the background?

Yes, long code generation processes can be handled in the background using background mode. This allows developers to monitor progress and continue other work while the AI agent completes complex coding tasks.

What is the best way to refactor code using multi-model AI agents?

The best way to refactor code with multi-model AI agents is delegating the task via command line, allowing the agent to process the request across multiple LLMs and select the optimal refactored output for your codebase.

Do I need a specific CLI tool and API key to use multi-model coding automation?

Yes, you need the Blackbox AI CLI installed and a valid API key to use this multi-model coding automation. These prerequisites authenticate your session and enable the agent to execute coding tasks.

Does the coding automation agent support multimodal image analysis?

Yes, the coding automation agent supports multimodal image analysis by automatically detecting images in your input. It switches to vision support mode to analyze visual data alongside text prompts.

Can I pause and resume interactive coding sessions during development?

Yes, you can pause and resume interactive coding sessions during development using the checkpoint and resume feature. This allows efficient workflow management by saving task states for later continuation.