One-click install
npx skills add https://github.com/swingerman/disciplined-agentic-engineering --skill feature-init
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: feature-init
Source: https://github.com/swingerman/disciplined-agentic-engineering/tree/main/engineer/skills/feature-init
Command: npx skills add https://github.com/swingerman/disciplined-agentic-engineering --skill feature-init

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill creates the correct starting structure for a new Disciplined Agentic Engineering (DAE) feature so the pipeline can proceed from a well-formed Ready contract and tracked work, with guardrailed setup for slug, numbering, branch, and validation metadata.

Core Features & Use Cases

  • New feature initialization: Creates a feature directory layout with the required feature.md (Foundation Design schema) without pre-generating downstream artifacts like progress/spec/plan.
  • From-discuss, standalone, and onboarding intake modes: Supports initializing from a feature_intake payload, running an intake interview when no payload is present, and reverse-engineering existing features during onboarding.
  • Guardrailed validation and routing: Enforces slug rules, autonomy/status constraints, relevant ADR existence checks, prevents collisions by redirecting to the right follow-up skill (feature-edit/promote/done rejection), and surfaces scope decomposition guidance.

Quick Start

Ask the AI to initialize a new feature for the DAE pipeline by creating the feature folder, feature.md, branch, and tracker entry.

Frequently Asked Questions about feature-init

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

FAQPage Schema
How do I initialize a new feature workspace for Disciplined Agentic Engineering?

To initialize a DAE feature workspace, this Skill creates the required feature folder, Ready contract metadata, branch, and tracker record. It enforces slug rules and validation to ensure your pipeline starts from a well-formed, tracked state.

What is a Ready contract in feature initialization?

A Ready contract is the required starting metadata structure created during feature initialization. It ensures your DAE pipeline has a well-formed foundation before any downstream generation like progress, spec, or session logs occurs.

Can I onboard an existing feature into the DAE pipeline?

Yes, you can onboard an existing feature into the DAE pipeline by reverse-engineering it into a consistent pipeline-ready state. The Skill handles intake from existing work, applies slug validation, and sets up the tracker record without pre-generating downstream artifacts.

How do I handle branch collisions when creating a new feature?

When branch or slug collisions occur during feature initialization, the Skill prevents conflicts by applying promote and redirect rules. It routes you to the correct follow-up skill like feature-edit or feature-promote to resolve the issue safely.

What is the best way to start greenfield work from a feature intake payload?

The best way to start greenfield work from a feature intake payload is to use this Skill to run an intake interview if no payload exists, or directly initialize the feature folder, branch, and tracker record if the payload is present.