teach-impeccable

Gathers project design context through codebase analysis and user questions, then persists guidelines to CLAUDE.md.

Updated Mar 4, 2026
One-click install
npx skills add https://github.com/gmolike/Claude-Template --skill teach-impeccable-gmolike
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: teach-impeccable
Source: https://github.com/gmolike/Claude-Template/tree/main/.agents/skills/teach-impeccable
Command: npx skills add https://github.com/gmolike/Claude-Template --skill teach-impeccable-gmolike

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI coding assistants lack awareness of your project's design intent, so every session produces inconsistent UI decisions. This Skill runs a one-time setup that captures your brand personality, aesthetic direction, and accessibility requirements, then saves them as persistent design guidelines for all future sessions. ## Core Features & Use Cases - Codebase Discovery: Scans README, package.json, existing components, brand assets, and CSS variables to infer existing design patterns before asking anything. - Targeted Questioning: Asks only about what could not be inferred, covering users, brand personality, aesthetic preferences, and accessibility needs. - Persistent Design Context: Writes a structured Design Context section (users, brand personality, aesthetic direction, design principles) into CLAUDE.md so every future session follows the same guidelines. - Use Case: A team starting a new React app runs this once; afterward, every AI-generated component automatically respects the established dark-mode palette, minimal aesthetic, and WCAG requirements. ## Quick Start Ask the assistant to run the teach-impeccable setup to establish design guidelines for this project.

Frequently Asked Questions about teach-impeccable

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

FAQPage Schema
How do I give my AI coding assistant persistent design guidelines?

Run a one-time design context setup that explores your codebase, asks targeted questions about brand and users, and writes a Design Context section to CLAUDE.md. Every future session then reads these guidelines automatically.

What is a Design Context section in CLAUDE.md?

It is a Markdown section documenting your users, brand personality, aesthetic direction, and 3-5 design principles. The AI reads it at session start so all generated UI work follows consistent design decisions.

Does the setup ask questions it could answer from the codebase?

No. It first scans the README, package.json, existing components, brand assets, and CSS variables, then only asks about gaps it could not infer, such as emotional goals or anti-references.

Can I update the design guidelines after the initial setup?

Yes. If CLAUDE.md already exists, the setup appends or updates the Design Context section rather than overwriting the file, so you can refine guidelines as the brand evolves.

When should I not use a one-time design context setup?

Skip it for throwaway prototypes or projects with no UI surface. It is designed for projects where consistent visual direction matters across many future AI-assisted sessions.