kernic for AI agents

Apps built by AI often look the same because the agent makes up new colors as it goes. Give it your system instead. One command and one rule later, everything it writes uses your tokens.

The one-command setup

kernic context midnight-neon -o .   # writes design.md + tokens.json

design.md carries explicit adherence rules plus every token; the adjacenttokens.json is W3C DTCG-style, so Figma, Terrazzo and Style Dictionary pipelines read it too.

Point the agent at it

One line in AGENTS.md, CLAUDE.md or .cursorrules:

Visual design: follow ./design.md exactly.
Use only its tokens. Never invent raw hex values.

Or go live over MCP

claude mcp add kernic -- npx kernic mcp

• list_systems: discover saved systems
• get_system: full spec of one system
• get_tokens: tokens in design-md, css, tailwind, fonts, dtcg or json format
• list_vibes: available vibe presets

Why it holds up

Token files use a version number and an extensions block. New features (motion, shadows, multi-brand) get added without breaking your existing files. Your systems are local JSON files, and they stay that way.