From lylacore
Use when working with agent identity systems, Mind definitions, or Soul Packages
How this skill is triggered — by the user, by Claude, or both
Slash command
/lylacore:lylacoreThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill when:
Use this skill when:
const { loadMind, loadMindsFromDirectory, validateMind } = require('lylacore/sdk/mind-loader');
// Load and validate a single Mind
const mind = loadMind('./minds/lilith.json');
// Load all Minds from a directory
const minds = loadMindsFromDirectory('./minds/');
// Validate Mind data
const result = validateMind(mindData, schema);
if (!result.valid) {
console.error(result.errors);
}
const { exportSoulPackage, importSoulPackage, validateSoulPackage } = require('lylacore/sdk/soul-package');
// Export Minds as a Soul Package
const pkg = exportSoulPackage(minds, {
author: 'Joysusy, Violet Klaudia',
description: 'Violet\'s Mind System',
tags: ['violet', 'identity']
});
// Import a Soul Package
const { minds, metadata } = importSoulPackage(pkg, {
selectMinds: ['Lilith', 'Rune'] // Optional: select specific Minds
});
// Validate package structure
const validation = validateSoulPackage(pkg);
const { learnPattern, applyStyle, analyzeStyle } = require('lylacore/sdk/coach-engine');
// Learn from user interaction
const styleMetadata = learnPattern(
userMessage,
agentResponse,
{ language: 'en', topic: 'coding' },
existingStyle
);
// Apply learned style to a message
const styledMessage = applyStyle(message, styleMetadata);
// Analyze communication style from message history
const profile = analyzeStyle(userMessages);
Defines the structure of a Mind instance:
Portable container for Mind instances:
Contextual Observation and Adaptive Communication Harmonization:
Lylacore is Layer 0 foundation — agent-agnostic capability engine:
See docs/LAYERED_IDENTITY_ARCHITECTURE.md for full design philosophy.
const { validateMind, loadSchema } = require('lylacore/sdk/mind-loader');
const mindData = {
name: 'Lilith',
symbol: '🌙',
role: 'Soul Package Architect',
traits: { precision: 0.9, creativity: 0.7 },
triggers: [
{
context_pattern: 'encryption|security|soul package',
activation_weight: 0.9
}
]
};
const schema = loadSchema();
const result = validateMind(mindData, schema);
if (result.valid) {
console.log('✓ Mind definition is valid');
} else {
console.error('✗ Validation errors:', result.errors);
}
const { loadMindsFromDirectory } = require('lylacore/sdk/mind-loader');
const { exportSoulPackage } = require('lylacore/sdk/soul-package');
const fs = require('fs');
// Load all Minds
const minds = loadMindsFromDirectory('./data/minds/');
// Create Soul Package
const pkg = exportSoulPackage(minds, {
author: 'Joysusy, Violet Klaudia',
description: 'Violet\'s complete Mind system',
tags: ['violet', 'production', 'v1.0']
});
// Save to file
fs.writeFileSync('./violet-soul-package.json', JSON.stringify(pkg, null, 2));
const { learnPattern, applyStyle } = require('lylacore/sdk/coach-engine');
let userStyle = null;
// Learn from each interaction
function onUserMessage(userMsg, agentResponse) {
userStyle = learnPattern(
userMsg,
agentResponse,
{ language: 'en', topic: 'development' },
userStyle
);
// Save userStyle to persistent storage
saveStyleMetadata(userStyle);
}
// Apply learned style to agent responses
function styleResponse(message) {
if (!userStyle) return message;
return applyStyle(message, userStyle);
}
// In lylacore/adapters/lavender-adapter.js
const { loadMind } = require('../sdk/mind-loader');
function enhanceMemoryWithIdentity(memory, activeMind) {
return {
...memory,
mindContext: activeMind.name,
mindSymbol: activeMind.symbol,
activationWeight: activeMind.triggers[0]?.activation_weight || 0.5
};
}
// VioletCore uses Lylacore SDK to load Violet's specific Minds
const { loadMindsFromDirectory } = require('lylacore/sdk/mind-loader');
const { decrypt } = require('violet-core/scripts/soul-cipher');
// Decrypt and load Violet's Minds
const encryptedData = fs.readFileSync('./data/minds-index.enc');
const decryptedData = decrypt(encryptedData, process.env.VIOLET_SOUL_KEY);
const minds = JSON.parse(decryptedData);
// Validate against Lylacore schema
minds.forEach(mind => {
const result = validateMind(mind, loadSchema());
if (!result.valid) throw new Error(`Invalid Mind: ${result.errors}`);
});
schemas/mind-v1.json — Mind Schema specificationdocs/LAYERED_IDENTITY_ARCHITECTURE.md — System design philosophyexamples/example-mind.json — Sample Mind definitionadapters/lavender-adapter.js — Memory system integrationAuthors: Joysusy & Violet Klaudia 💖
npx claudepluginhub joysusy/violet-plugin-place --plugin lylacoreRebuilds agent identity after cold starts: progressive context loading, fresh-vs-continuation detection, calibration, and coherence verification. Use at session start or after interruptions.
Knowledge base on Claude Code formats, patterns, and configurations for commands, agents, skills, hooks, memory, plugins, settings. Use for creating, improving, auditing components.
Design agent-native architectures where agents are first-class citizens. Covers parity between UI and agent capabilities, atomic tools, and building autonomous loops.