# canonical skill-map prototype — Gemini + Codex

A deployed VibeSkill prototype that combines Gemini-generated frontend scaffolding with Codex-led structural refinement, replacing placeholder branches with canonical skill mappings and converting the sidebar into a denser evidence-oriented skill dossier.

Type: project entry

Published: 2026-03-31

[Original page](<https://artlu.ai/project/canonical-skill-map-prototype-gemini-codex>)

[Browse this group](<https://artlu.ai/ai/groups/other.md>)

Parent project: VibeSkill

Recorded status: launched

Updated: 2026-09-16T16:22:22.597Z

Technologies: React, Vite, Tailwind CSS, motion/react, lucide-react, Cloudflare Pages, Codex, Gemini

## Details

Built and deployed a public VibeSkill prototype that bridges generative UI prototyping and structured product engineering. The initial frontend and interaction model came from Gemini, while Codex was used to refine the prototype into a more accurate software-engineering skill map: placeholder child nodes were replaced with canonical first-level domain branches, child expansion geometry was corrected for better spatial consistency, and the right-hand detail panel was reshaped into a denser documentation and evidence-oriented skill dossier. The result is a live prototype that preserves the approved neural-map landing screen while moving the underlying content model closer to an objective, reference-driven representation of software engineering knowledge.

## Links

[Site or demo](<https://vibeskill-prototype.pages.dev/>)

## Related build notes

[launched the canonical VibeSkill prototype](<https://artlu.ai/ai/log/day-14-launched-the-canonical-vibeskill-prototype.md>)


Status and dates describe the published record, not a recent test of an external service. Missing details are not inferred. Public links do not imply an open-source licence or permission to reuse material.

[Directory](<https://artlu.ai/llms.txt>)
