todo-dsl-generator

Convert rough proposal text into a structured TODO DSL plan with step metadata and references.

174|17|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/CurryTang/Amadeus --skill todo-dsl-generator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: todo-dsl-generator
Source: https://github.com/CurryTang/Amadeus/tree/main/skills/todo-dsl-generator
Command: npx skills add https://github.com/CurryTang/Amadeus --skill todo-dsl-generator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converting vague ideas, proposals, or chat transcripts into a precise, executable sequence of steps with clear ownership and traceable references.

Core Features & Use Cases

  • Transforms rough input into a structured TODO DSL with step metadata (step_id, title, kind, objective, assumptions, acceptance, commands, checks, depends_on, references).
  • Includes per-step knowledge and codebase references to ensure low-bias orchestration and easy traceability.
  • Supports exporting flattened todoCandidates for project management views and Kanban-like workflows.

Quick Start

Provide a rough idea or proposal text and your project metadata, and request a structured TODO DSL with per-step references.

Frequently Asked Questions about todo-dsl-generator

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

FAQPage Schema
How do I convert rough project ideas into structured executable steps?

Converting rough project ideas into structured executable steps requires parsing proposal text into a TODO DSL plan. This format defines ordered steps with objectives, acceptance criteria, and traceable knowledge references to ensure low-bias orchestration.

What is a TODO DSL and when do I need it for feature planning?

A TODO DSL is a structured domain-specific language for product planning that outputs an object with ordered steps. You need it when feature proposals require clear, testable steps with per-step knowledge and codebase references for traceability.

How do I structure task dependencies and acceptance criteria from chat transcripts?

Structuring task dependencies from chat transcripts involves generating a plan with step_id, depends_on, and acceptance fields. This validates proposal text into ordered steps with explicit assumptions and commands for testable execution.

Can I export structured plans into a flattened array for Kanban workflows?

You can export structured plans into a flattened todoCandidates array for Kanban-like workflows. This output transforms the ordered TODO DSL steps into a project management view for easy tracking and execution.

Does this task automation approach validate steps against existing knowledge assets?

This task automation approach validates generated steps against existing knowledge assets and the project codebase. It attaches per-step knowledgeReferences and codebaseReferences to ensure low-bias orchestration and easy traceability.