vibe-coding

Porkbutts n' Taters

An opinionated collection of skills to add structured workflows to vibecoding.

Inspired by Cow and Chicken's love of porkbutts n' taters.

Installation

From GitHub

/plugin marketplace add Porkbutts/porkbutts-n-taters
/plugin install vibe-coding@porkbutts-n-taters

For Local Development

/plugin marketplace add /path/to/this/repo
/plugin install vibe-coding@porkbutts-n-taters

Then restart Claude Code.

Plugins

vibe-coding

Personal workflow for getting better results out of vibecoding.

Skills

product-design - Design MVPs through adaptive interviews that achieve buildable clarity. Guides you through interviewing to extract implementation-ready requirements, then produces a PRD.

• Triggers when you want to define, scope, or document an application idea
• Interview phases: Core Concept, User Scenarios, Technical Context, Scope Boundaries
• Output: Structured PRD with roles, scenarios, acceptance criteria, and technical context

architecture - Design technical architecture from a PRD. Transform product requirements into implementation-ready technical specs.

• Use when you want to create an architecture document, define system design, or plan technical implementation
• Reads from docs/PRD.md and outputs docs/ARCHITECTURE.md
• Includes: tech stack decisions with rationale, directory structure, key abstractions, data flow, and anti-patterns to avoid

functional-prototype - Build working prototypes from PRDs with mock data and stubbed integrations to validate flows before full implementation.

• Triggers: "build a prototype", "functional prototype", "make a demo", "implement this PRD", "prototype this spec"
• Output: Working clickable prototype; optionally generates docs/integration-guide.md (asks first) documenting how to replace stubs with real integrations

task-decomposition - Break down PRD and Architecture into implementable engineering tasks.

• Use when you want to create a task list, plan implementation work, or decompose features into stories and tasks
• Reads from docs/PRD.md and docs/ARCHITECTURE.md, outputs docs/TASKS.md
• Structure: Epics → User Stories → Atomic Tasks with acceptance criteria and test approaches

bootstrap-project - Bootstrap a project with CI/CD, pre-commit hooks, CLAUDE.md, and framework best practices.

• Triggers: "bootstrap project", "setup CI", "add pre-commit hooks", "add husky", "project setup", "init CLAUDE.md"
• Ensures GitHub CI workflow exists for lint, typecheck, test with coverage, and build
• Posts coverage summary comments on PRs via vitest-coverage-report-action or jest-coverage-comment
• Sets up husky + lint-staged pre-commit hooks for lint, format, and typecheck
• Adds Claude Code hook guard to block --no-verify and prevent agents from bypassing pre-commit hooks
• Generates a CLAUDE.md with project overview, architecture summary, key commands, code conventions, and rules — derived from app description, docs, and file structure
• Installs Vercel best practices via npx add-skill vercel-labs/agent-skills
• Enforces high code coverage thresholds (80%) appropriate for LLM coding agents
• Detects existing config and only adds what's missing

task-implementation - Implement a task end-to-end: read task spec, create git worktree, write tests (TDD), implement feature, verify coverage thresholds, update docs, push branch, and create PR.

• Triggers: "implement task 1.2.3", "work on the auth task", "build task 3.1.1"
• Reads task specs from docs/tasks/task-<id>.md (falls back to docs/TASKS.md)
• Reads docs/PRD.md, docs/ARCHITECTURE.md for context; docs/BRAND-GUIDELINES.md for UI work
• Strict TDD: writes failing tests first, then implements minimum code to pass
• Enforces 80% coverage thresholds (statements, branches, functions, lines)
• Handles review feedback loops (reads ## Review Feedback section from task spec)
• Ships: commits, pushes feature branch, creates PR with gh

review-pr - Review pull requests against task specifications.

• Triggers: "review PR", "code review", "is this ready to merge", "check my PR"
• Reads task spec and branch diff, verifies acceptance criteria, audits test coverage
• Checks for correctness, security (OWASP top 10), architectural adherence, and scope creep
• Output: Structured verdict (APPROVE / REQUEST CHANGES) with evidence per criterion and a Manual QA Checklist
• Posts review findings as a comment on the PR via gh pr comment

brand-generator - Generate unique, ownable brand names with verified-available domains through structured discovery and linguistic construction.