feature-implementation

Orchestrates end-to-end feature implementation through a six-phase WHY/WHAT/HOW workflow with quality gates.

4|2|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/Arete-Consortium/ai-skills --skill feature-implementation-arete-consortium
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feature-implementation
Source: https://github.com/Arete-Consortium/ai-skills/tree/main/workflows/feature-implementation
Command: npx skills add https://github.com/Arete-Consortium/ai-skills --skill feature-implementation-arete-consortium

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Implementing a multi-file feature without structure leads to missed requirements, regressions, and undocumented changes. This Skill enforces a disciplined pipeline — context mapping, requirements capture, design, test-first implementation, verification, and PR delivery — so every feature ships with clear intent and passing quality gates. ## Core Features & Use Cases - Six-Phase Pipeline: Runs context mapping, WHY requirements, WHAT design, HOW implementation, verification, and PR delivery as structured phases with artifacts. - Blocking Quality Gates: Enforces tests, lint, type check, and build after implementation, blocking the pipeline on failure. - Escalation and Self-Correction: Defines retry limits per error class and escalates ambiguous requirements or design conflicts to the user. - Use Case: When asked to add a new API endpoint spanning routes, handlers, types, and tests, the workflow maps the codebase first, validates requirements with you, implements test-first, and opens a documented pull request. ## Quick Start Ask the agent to implement a complete feature in your repository using the feature-implementation workflow, providing the feature description and the absolute path to your codebase.

Frequently Asked Questions about feature-implementation

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

FAQPage Schema
How do I implement a feature across multiple files with an AI agent?▼

Use a phased workflow that maps the codebase first, captures requirements, defines a design, then implements test-first with quality gates. This Skill runs six phases from context mapping through PR creation, producing structured artifacts at each step.

What is the WHY/WHAT/HOW framework for feature development?▼

WHY captures the goal, motivation, success criteria, and anti-goals. WHAT defines architecture, files to create or modify, interfaces, and data flow. HOW executes implementation steps with acceptance criteria and quality gates between steps.

When should I not use a full feature implementation workflow?▼

Skip it for small bug fixes or single-file changes, where the pipeline overhead exceeds the change complexity. Use an engineering persona directly, a context-mapping workflow for reconnaissance only, or a planning persona if you need a plan without execution.

Does the workflow push code directly to the main branch?▼

No. The delivery phase always creates a feature branch and a pull request with WHY/WHAT/HOW documentation. Pushing directly to main is explicitly prohibited, and violations trigger self-correction that reverts and creates a PR instead.

What happens when tests fail during implementation?▼

New test failures trigger up to three fix-and-rerun retries. Regressions in existing tests stop the pipeline after one retry for root-cause analysis. Repeated failures beyond the retry limit halt the workflow and report what was attempted.