blackbox

Automate multi-model coding task delegation and evaluation via the Blackbox CLI.

78|16|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill blackbox-sheawinkler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/sheawinkler/hermes-agent-ultra/tree/main/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill blackbox-sheawinkler

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the delegation and evaluation of coding tasks across multiple AI models via a CLI, reducing turnaround time and improving solution quality for software projects.

Core Features & Use Cases

  • Multi-model coding agent: dispatches tasks to Claude, Codex, Gemini, Blackbox Pro, and uses a built-in judge to select the best implementation.
  • Flexible workflows: supports interactive sessions, one-shot prompts, and checkpoint-resume for long-running tasks.
  • Safe governance: policy-based tool use with session management, MCP integration, and auditing for production-grade workflows.

Quick Start

Ask Blackbox to scaffold a REST API endpoint with authentication.

Frequently Asked Questions about blackbox

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

FAQPage Schema
How do I delegate coding tasks to multiple AI models simultaneously?

Multi-model task delegation dispatches coding prompts to models like Claude, Codex, and Gemini simultaneously, using a built-in judge to evaluate and select the best implementation.

Can I resume long-running coding tasks if an interactive session is interrupted?

Checkpoints allow resuming long-running coding tasks after interruptions. The system supports checkpoint-resume alongside interactive sessions and one-shot prompts to maintain continuous workflow execution.

Do I need an API key to configure providers for multi-model coding agents?

A valid API key is required to configure providers and models for multi-model coding agents. The Blackbox CLI must also be installed to automate task delegation and evaluation.

What is the best way to evaluate AI-generated code across different platforms?

Parallel model testing via a CLI evaluates AI-generated code across platforms. A built-in judge reviews implementations from different models to select the highest quality solution.

Does multi-model task delegation support policy-based tool use and session auditing?

Multi-model task delegation supports safe governance through policy-based tool use, session management, MCP integration, and auditing to ensure production-grade workflows for software projects.

When should I not use a multi-model coding agent for software projects?

Avoid multi-model coding agents for simple, single-solution tasks that do not require cross-model evaluation. The overhead of dispatching to multiple models and judging results is best suited for complex coding challenges.