feature

Automate feature implementation from planning to GitHub pull request creation.

Updated Aug 14, 2012
One-click install
npx skills add https://github.com/dakotahp/dotfiles --skill feature-dakotahp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: feature
Source: https://github.com/dakotahp/dotfiles/tree/main/dot_config/claude/skills/feature
Command: npx skills add https://github.com/dakotahp/dotfiles --skill feature-dakotahp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a disciplined, end-to-end pipeline for implementing new software features with test-driven development, automated verification, and formal code review, removing ad hoc approaches that lead to unstable or unreviewed changes.

Core Features & Use Cases

  • Structured Planning & Spec: Runs brainstorming and plan-writing subroutines to produce a committed spec and implementation plan.
  • TDD-First Implementation: Produces falsifiable prove statements, writes failing tests, and drives implementation via dispatched subagents that commit to a feature branch.
  • Verification, Simplification, and PR Flow: Runs code simplification, records prove_it evidence, performs formal code review, and creates a draft GitHub pull request, then iterates on review feedback until approvals are obtained.

Quick Start

Invoke the pipeline by telling the agent to run the feature command with your description, for example: /feature add password reset flow with 1-hour expiry tokens

Frequently Asked Questions about feature

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

FAQPage Schema
How do I automate test-driven development from a feature spec to a GitHub PR?

You can automate test-driven development by orchestrating a pipeline that generates a plan, writes failing tests, implements code via subagents, verifies with prove_it, and creates a GitHub pull request. This removes ad hoc approaches and unreviewed changes.

What is the best way to automate code review and PR creation for new features?

Automating code review and PR creation involves running code simplification, recording verification evidence, and performing formal reviews before creating a draft GitHub pull request. The pipeline iterates on review feedback until formal approvals are obtained.

Do I need the gh CLI and prove_it to run an automated TDD feature pipeline?

Yes, you need the gh CLI and prove_it available on your PATH to run the automated TDD feature pipeline. You also need a git repository with a non-protected feature branch and the ability to run project-specific tooling like tests and linters.

How does a subagent-based workflow handle feature implementation and verification?

A subagent-based workflow handles feature implementation by dispatching agents to write failing tests and drive code commits to a feature branch. Verification is handled by recording prove_it evidence and running code simplification before formal review.

Can I use automated TDD pipelines with Convext or Slack integrations?

Yes, automated TDD pipelines support optional Convext or Slack integrations during the end-to-end feature implementation process. These integrations work alongside the planning, test-driven development, and GitHub pull request creation stages.

When should I avoid using an autonomous TDD feature pipeline?

You should avoid using an autonomous TDD feature pipeline if you lack a non-protected feature branch, do not have the gh CLI or prove_it on your PATH, or cannot install and run project-specific tooling to execute tests and linters.