quaks-generator

Generate quaks agent scaffolds with SupervisedWorkflowAgentBase and graph patterns.

1|Updated Oct 13, 2025
One-click install
npx skills add https://github.com/bsantanna/quaks --skill quaks-generator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quaks-generator
Source: https://github.com/bsantanna/quaks/tree/main/.claude/skills/quaks-generator
Command: npx skills add https://github.com/bsantanna/quaks --skill quaks-generator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Create and register new autonomous agents in quaks by generating standardized scaffolds and boilerplate code that wire them into the system.

Core Features & Use Cases

  • Scaffolds a new agent prototype that inherits from SupervisedWorkflowAgentBase and registers it within the quaks toolkit.
  • Supports two graph patterns — direct graph (deterministic edges) and command multi-agentic (LLM-based routing) — to cover fixed pipelines and dynamic workflows.
  • Outputs ready-to-use code structure (agent class, state, prompts, and wiring) plus documentation to accelerate deployment in a project.

Quick Start

Generate a new agent scaffold in a specified directory, choose a graph pattern, and register it with the agent registry.

Frequently Asked Questions about quaks-generator

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

FAQPage Schema
How do I scaffold a new autonomous agent in LangGraph?

To scaffold a new autonomous agent in LangGraph, you specify a target directory and intended roles to generate boilerplate code, state definitions, prompts, and configuration wiring that registers the agent within your project.

What is the difference between deterministic edges and LLM-based routing for multi-agent workflows?

Deterministic edges create a fixed direct graph pipeline with predictable execution paths, while LLM-based routing uses a command multi-agentic pattern to dynamically direct workflow steps based on language model decisions.

How do I generate boilerplate code for a supervised multi-agent workflow?

You generate boilerplate code by specifying a directory, defining agent roles, and selecting a graph pattern to produce ready-to-register code skeletons that inherit from SupervisedWorkflowAgentBase.

Can I prototype a new LangGraph agent without manually writing configuration wiring?

Yes, you can prototype a new LangGraph agent without manual configuration wiring because this generator outputs complete code skeletons, prompts, and wiring that automatically integrate with the agent registry.

Do I need to install dependencies to use the SupervisedWorkflowAgentBase for agent scaffolding?

The agent scaffolding process requires integration with the SupervisedWorkflowAgentBase, but the generator itself operates without external dependencies to produce the necessary code structure and wiring for deployment.