auto-author-workflow

Compile natural language descriptions into validated JSON workflow topologies.

Updated Jun 12, 2026
One-click install
npx skills add https://github.com/shawnroos/auto --skill auto-author-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-author-workflow
Source: https://github.com/shawnroos/auto/tree/main/skills/auto-author-workflow
Command: npx skills add https://github.com/shawnroos/auto --skill auto-author-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill removes the complexity of manually writing JSON configurations for agentic loop topologies, ensuring that custom workflows are syntactically correct and logically sound before execution.

Core Features & Use Cases

  • Prose-to-Workflow Compilation: Converts natural language descriptions of loop topologies into validated JSON workflow files.
  • Reverse-Derivation: Extracts and upgrades the topology of a completed run from its run-record, allowing users to save successful processes as reusable templates.
  • Strict Validation: Uses built-in linting and topology rendering to ensure every workflow meets the engine's structural requirements before it is saved to the workspace or global tier.

Quick Start

Ask the auto-author-workflow skill to create a new workflow that runs three parallel research steps followed by a single synthesis step.

Frequently Asked Questions about auto-author-workflow

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

FAQPage Schema
How do I create agentic workflows from natural language descriptions?

The auto loop engine requires strict structural validation and atomic file writing to execute workflows. This process enforces those structural requirements before saving topologies to your workspace.

Can I turn completed run-records into reusable workflow templates?

Yes, reverse-derivation extracts and upgrades the topology of a completed run from its run-record. This allows you to save successful agentic processes as reusable workflow templates.

How do I generate JSON loop topologies without manual coding errors?

Compiling plain-language descriptions into validated JSON loop topologies removes manual coding errors. Built-in linting and topology rendering ensure custom workflows are syntactically correct and logically sound before execution.

Does the auto loop engine validate workflow topology before saving files?

Yes, strict validation uses built-in linting and topology rendering to ensure every workflow meets the engine's structural requirements. Atomic file writing guarantees compatibility with the execution contract before saving.

What is the best way to define parallel steps in an agentic loop topology?

Describing your desired steps in natural language allows the compiler to generate the validated JSON workflow topology. You can request parallel research steps followed by a single synthesis step.