The Antidote
Every AI model learned from the same templates. Without intervention, they all produce the same predictable mistakes. Impeccable names them, detects them, and teaches the AI to avoid them.
25 patterns are caught automatically by the CLI and browser overlay -- no LLM needed. The rest are caught by /critique through design analysis.
npx impeccable detect src/
The Framework
One skill with deep design expertise. 20 commands that form the vocabulary of design. And a CLI that catches what humans miss.
Get Started
Two steps to impeccable design.
npx skills add pbakaus/impeccable
Other install methods
/plugin marketplace add pbakaus/impeccable
/plugin to install from Discover tab
Keep the command cheatsheet handy for quick reference. To update skills, run npx skills update.
What's New
- Anti-pattern detection engine: 25 rules across typography, color, layout, visual details, motion, and design quality
- CLI tool:
npx impeccable detect [target]for standalone scanning outside AI harnesses - Browser visualizer: live overlays highlighting detected anti-patterns directly on the page
/critiqueoverhaul: sub-agent architecture with scoring, personas, and cognitive load assessment- Gallery of Shame: real examples with live detection overlays
- New provider: Trae (China + International)
/critiquenow scores against Nielsen's 10 heuristics, tests with persona archetypes, and assesses cognitive load/auditnow scores 5 dimensions with P0-P3 severity ratings and structured action plans- Improved skill descriptions for better agent auto-discovery
- Fixed invalid YAML frontmatter that broke GitHub preview and Codex loading (#67)
- Codex CLI now uses correct
$prefix for command references
View older releases
/typesetnow recommends fixed type scales for app UIs, reserving fluid typography for marketing/content pages
- 3 new skills:
/typeset(fix typography),/arrange(fix layout & spacing),/overdrive(technically extraordinary effects, beta) - Skills now auto-gather design context via
.impeccable.md— run/teach-impeccableonce, all skills benefit - Deep linking to commands (
#cmd-overdrive, etc.)
- Added OpenCode provider support
- Added Pi provider support
- Recategorized
/onboardas an enhancement command
- Added Kiro support (
.kiro/skills/) - Restored prefix toggle — download
i-prefixed bundles to avoid naming conflicts - Audit and critique skills only suggest real, installed commands
- Unified skills architecture — commands are now skills with
user-invocable: true - Added VS Code Copilot and Google Antigravity support (
.agents/skills/) - New install flow:
npx skills addas primary, universal ZIP as fallback - Added universal ZIP containing all 5 provider directories
- Renamed
/simplifyto/distillto avoid Claude Code conflict
- Initial release with enhanced frontend-design skill
- 17 design commands: /polish, /audit, /distill, /bolder, and more
- Support for Cursor, Claude Code, Gemini CLI, and Codex CLI
- Interactive command cheatsheet
Frequently Asked Questions
Where do I put the downloaded files?
The easiest way is npx skills add pbakaus/impeccable — it auto-detects your AI harness and places files correctly.
If you downloaded the universal ZIP, extract it to your project root (same level as your package.json or src/ folder). It creates hidden folders for each supported tool: .cursor/, .claude/, .gemini/, .codex/, and .agents/.
Project-level installation takes precedence and lets you version control your skills.
How do I update to the latest version?
- All tools:
npx skills update(ornpx skills checkto see what's new first) - Claude Code plugin: Open
/plugin, go to the Discover tab, and update from there - Manual ZIP: Download the latest ZIP from above and extract to your project root, overwriting existing files
Your .impeccable.md design context file (created by /impeccable teach) is never overwritten — your project context is preserved across updates.
Commands or skills aren't appearing. What do I do?
For commands: Type / in your AI harness and look for commands like /audit, /polish, etc. If they don't appear, double-check the files are in the correct location.
For skills: Skills are applied automatically when relevant. To verify, explicitly mention "use the impeccable skill" in your prompt—this forces the AI to acknowledge and apply it.
Tool-specific setup:
- Cursor: Requires Nightly channel + Agent Skills enabled in Settings → Rules
- Gemini CLI: Requires
@google/gemini-cli@preview+ Skills enabled via/settings
I'm new to AI harnesses. Where do I start?
Skills and commands are intermediate features. If you're just getting started, learn the basics first:
- Claude Code: Official Documentation
- Cursor: Cursor Docs
- Gemini CLI: Gemini CLI Docs
- Codex CLI: Codex GitHub
Once you're comfortable with basic prompting and code generation, come back and give Impeccable a try.
Work with me
I help teams navigate AI transformation: upleveling developers, building AI-native workflows, and shipping products. Deep expertise in developer experience, design systems, and full-stack architecture.