blackbox

Delegate coding tasks to the Blackbox CLI agent with multi-model adjudication.

Updated Mar 25, 2026
One-click install
npx skills add https://github.com/afel6/scal-ai-pipeline --skill blackbox-afel6
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/afel6/scal-ai-pipeline/tree/main/hermes_skills_library/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/afel6/scal-ai-pipeline --skill blackbox-afel6

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegate coding tasks to the Blackbox CLI to reduce manual implementation work and get evaluated, higher-quality code by running multiple models and selecting the best result.

Core Features & Use Cases

  • Multi-model adjudication: Runs the same prompt across multiple LLMs and uses a judge to pick the best implementation.
  • Interactive and background workflows: Supports PTY interactive sessions, long-running background tasks, polling, logs, checkpoints, and resume.
  • PR reviews and refactoring: Safely review pull requests, refactor modules, add tests, and prototype features in isolated temp directories.
  • Multimodal support: Detects images and can switch visual language models for vision-enabled coding tasks.

Quick Start

Use the blackbox skill to refactor the auth module to OAuth 2.0 in my project directory.

Frequently Asked Questions about blackbox

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

FAQPage Schema
Can I run automated pull request reviews using multiple LLMs?

Interactive coding sessions require a PTY-capable terminal, Node.js 20+, the @blackboxai/cli installed, and a Blackbox AI API key configured. This environment supports long-running background tasks, polling, logs, checkpoints, and resume functionality.

Does multi-model code generation work with image inputs?

Multimodal coding tasks are supported by detecting images and switching visual language models for vision-enabled coding tasks. This allows the agent to process and implement code based on visual inputs.

What do I need to set up interactive coding sessions from a terminal?

Interactive coding sessions require a PTY-capable terminal, Node.js 20+, the @blackboxai/cli installed, and a Blackbox AI API key configured. This environment supports long-running background tasks, polling, logs, checkpoints, and resume functionality.

How does multi-model code generation improve refactoring results?

Multi-model code generation improves refactoring by running the same prompt across multiple LLMs and using a judge to select the best implementation, ensuring higher quality output. This mechanism reduces manual implementation work and evaluates code automatically.

Can I run automated pull request reviews using multiple LLMs?

Automated pull request reviews can be run using multiple LLMs by delegating tasks to the Blackbox CLI agent. It safely reviews pull requests, refactors modules, and adds tests in isolated temp directories.