skills.rest
Signmanal
SignmanalCOMMUNITY
Explore all skills by this author →
GitHub

blackbox

Delegate coding tasks to the Blackbox AI CLI agent for multi-model implementation evaluation.

1|Updated Jun 25, 2026
Search Tags:#code generation#task automation#pr review#multi-model#checkpoint#coding agent#blackbox ai
One-click install
npx skills add https://github.com/Signmanal/VIGIL --skill blackbox-signmanal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: blackbox
Source: https://github.com/Signmanal/VIGIL/tree/main/optional-skills/autonomous-ai-agents/blackbox
Command: npx skills add https://github.com/Signmanal/VIGIL --skill blackbox-signmanal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the tedious manual work of testing code implementations across multiple LLMs to identify the highest quality output, saving developers hours of trial and error when building or modifying code.

Core Features & Use Cases

  • Multi-Model Code Evaluation: Dispatches the same coding task to multiple LLMs (Claude, Codex, Gemini, Blackbox Pro) and uses a built-in judge to select the best implementation.
  • Flexible Task Execution: Supports one-shot quick tasks, background long-running jobs, checkpoint-based resumption for paused work, and automated PR reviews.
  • Use Case: A developer can use this Skill to generate a JWT authentication implementation for an Express API, automatically evaluate versions from multiple models, and receive the most secure, efficient result without manual comparison.

Quick Start

Use the blackbox skill to delegate the coding task of adding JWT authentication with refresh tokens to your Express API and receive the best multi-model evaluated implementation.

Frequently Asked Questions about blackbox

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

FAQPage Schema
How do I compare code generation results from multiple LLMs automatically?▼

To compare code generation results from multiple LLMs automatically, you can dispatch the same coding task to models like Claude, Codex, and Gemini. A built-in judge evaluates the outputs and selects the best implementation.

Can I use a coding agent to automate pull request reviews?▼

Yes, you can use a coding agent to automate pull request reviews. The agent evaluates code changes across multiple models to ensure secure and efficient refactoring without manual comparison.

Does multi-model code evaluation work for Node.js API implementations?▼

Multi-model code evaluation works for Node.js API implementations. It dispatches tasks like building JWT authentication to various LLMs and selects the most secure, efficient result for your Express projects.

How do I resume a long-running coding task after pausing it?▼

To resume a long-running coding task after pausing it, you use checkpoint-based resumption. This feature allows the agent to save progress and continue background execution from the last saved state.

What is the best way to automate bug fixes across different coding models?▼

The best way to automate bug fixes across different coding models is to delegate the task to an agent that dispatches the prompt to multiple LLMs. The built-in judge then selects the highest quality refactored output.

Do I need an interactive terminal session for multi-model code generation?▼

You do not need an interactive terminal session for multi-model code generation, but PTY-enabled interactive terminal session support is available. This allows for flexible execution alongside one-shot tasks and background jobs.

Related Skills

Explore More Software Engineering Skills→
lihaoze123lihaoze123
2

trellis-start

Initialize coding sessions by loading project context and routing workflows.

Community
Intermediate
tabgabtabgab

opp-repl-smoke-tests

Run opp_repl smoke tests to verify simulations start and terminate without crashing.

Community
Intermediate
AngelP17AngelP17
1

senior-devops

Automate CI/CD pipelines, infrastructure provisioning, and deployments across AWS, GCP, and Azure.

Community
Intermediate
cloudlinesscloudliness

blackbox

Dispatch coding tasks to multiple LLMs and select the best implementation.

Community
Advanced