character-generator

Character Generator Skill

An intelligent skill that creates production-ready elizaOS character configurations with comprehensive personality traits, knowledge bases, and plugin integrations.

When to Use

This skill activates when you need to:

• Create a new elizaOS character from scratch
• Generate character configurations for specific use cases
• Set up agent personalities and behaviors
• Configure multi-platform agent deployments

Trigger phrases:

• "Create a character for [purpose]"
• "Generate an elizaOS agent configuration"
• "Build a character that [does something]"
• "Set up an agent for [platform/use case]"

Capabilities

This skill can:

1. 🎭 Design character personalities and traits
2. 📚 Set up knowledge bases and training data
3. 🔌 Configure plugin ecosystems
4. 💬 Generate message examples and conversation patterns
5. 🎨 Define writing styles for different contexts
6. 🔐 Set up secure secrets management
7. 🌐 Configure multi-platform deployments
8. ✅ Validate character configurations

Workflow

Phase 1: Requirements Gathering

Ask these questions to understand the character:

1. Purpose: "What is the primary purpose of this agent?"

   • Customer support
   • Content creation
   • Technical assistance
   • Community management
   • Data analysis
   • Creative collaboration

2. Personality: "What personality traits should the agent have?"

   • Professional vs. Casual
   • Serious vs. Humorous
   • Concise vs. Detailed
   • Technical vs. Accessible

3. Knowledge Domain: "What topics should the agent be expert in?"

   • Programming languages
   • Business domains
   • Creative fields
   • Technical support areas

4. Platforms: "Which platforms will the agent operate on?"

   • Discord
   • Telegram
   • Twitter
   • Web interface
   • Custom integrations

5. Special Features: "Are there any special capabilities needed?"

   • Voice synthesis
   • Image generation
   • Web search
   • Database access
   • Custom actions

Phase 2: Character Design

Based on requirements, design the character structure:

interface CharacterDesign {
  // Core Identity
  name: string;              // Agent display name
  username: string;          // Platform username
  bio: string[];            // Personality description

  // Personality Traits
  adjectives: string[];     // Character traits
  topics: string[];         // Knowledge domains

  // Communication Style
  style: {
    all: string[];         // Universal rules
    chat: string[];        // Conversational style
    post: string[];        // Social media style
  };

  // Training Data
  messageExamples: Memory[][];  // Conversation examples
  postExamples: string[];       // Social post examples

  // Knowledge & Capabilities
  knowledge: KnowledgeItem[];   // Knowledge sources
  plugins: string[];            // Enabled plugins

  // Configuration
  settings: Settings;           // Agent settings
  secrets: Secrets;             // Environment variables
}

Phase 3: Implementation

Step 1: Create Character File

// characters/{name}.ts

import { Character } from '@elizaos/core';

export const character: Character = {
  // === CORE IDENTITY ===
  name: '{CharacterName}',
  username: '{username}',

  // Bio: Multi-line for better organization
  bio: [
    "{Primary role and expertise}",
    "{Secondary capabilities}",
    "{Personality traits}",
    "{Communication style}"
  ],

  // === PERSONALITY ===
  adjectives: [
    "{trait1}",
    "{trait2}",
    "{trait3}",
    "{trait4}",
    "{trait5}"
  ],

  topics: [
    "{topic1}",
    "{topic2}",
    "{topic3}",
    "{topic4}"
  ],

  // === COMMUNICATION STYLE ===
  style: {
    all: [
      "{Universal rule 1}",
      "{Universal rule 2}",
      "{Universal rule 3}"
    ],
    chat: [
      "{Chat-specific rule 1}",
      "{Chat-specific rule 2}",
      "{Chat-specific rule 3}"
    ],
    post: [
      "{Social media rule 1}",
      "{Social media rule 2}",
      "{Social media rule 3}"
    ]
  },

  // === TRAINING EXAMPLES ===
  messageExamples: [
    // Conversation 1: Greeting
    [
      {
        name: "{{user}}",
        content: { text: "Hello!" }
      },
      {
        name: "{CharacterName}",
        content: {
          text: "{Character's greeting response}"
        }
      }
    ],
    // Conversation 2: Main use case
    [
      {
        name: "{{user}}",
        content: { text: "{User question about primary use case}" }
      },
      {
        name: "{CharacterName}",
        content: {
          text: "{Detailed, helpful response showcasing expertise}"
        }
      },
      {
        name: "{{user}}",
        content: { text: "{Follow-up question}" }
      },
      {
        name: "{CharacterName}",
        content: {
          text: "{Continued helpful response}"
        }
      }
    ],
    // Conversation 3: Error handling
    [
      {
        name: "{{user}}",
        content: { text: "{Question outside expertise}" }
      },
      {
        name: "{CharacterName}",
        content: {
          text: "{Polite acknowledgment of limitation + redirect}"
        }
      }
    ]
  ],

  postExamples: [
    "{Example social post 1 showcasing personality}",
    "{Example social post 2 demonstrating expertise}",
    "{Example social post 3 showing communication style}"
  ],

  // === KNOWLEDGE ===
  knowledge: [
    "{Simple fact 1}",
    "{Simple fact 2}",
    {
      path: "./knowledge/{domain}",
      shared: true
    }
  ],

  // === PLUGINS ===
  plugins: [
    '@elizaos/plugin-bootstrap',
    '@elizaos/plugin-sql',

    // LLM Providers (conditional)
    ...(process.env.OPENAI_API_KEY ? ['@elizaos/plugin-openai'] : []),
    ...(process.env.ANTHROPIC_API_KEY ? ['@elizaos/plugin-anthropic'] : []),

    // Platform Integrations (conditional)
    ...(process.env.DISCORD_API_TOKEN ? ['@elizaos/plugin-discord'] : []),
    ...(process.env.TELEGRAM_BOT_TOKEN ? ['@elizaos/plugin-telegram'] : []),
    ...(process.env.TWITTER_API_KEY ? ['@elizaos/plugin-twitter'] : []),

    // Additional Capabilities
    {add_plugins_based_on_requirements}
  ],

  // === SETTINGS ===
  settings: {
    secrets: {},
    model: 'gpt-4',
    temperature: 0.7,
    maxTokens: 2000,
    conversationLength: 32,
    memoryLimit: 1000
  }
};

export default character;

Step 2: Create Knowledge Directory

mkdir -p knowledge/{name}

Create knowledge files:

• knowledge/{name}/README.md - Overview
• knowledge/{name}/core-knowledge.md - Domain expertise
• knowledge/{name}/faq.md - Common questions
• knowledge/{name}/examples.md - Use case examples

Step 3: Create Environment Template

# .env.example

# === LLM PROVIDERS ===
# OpenAI Configuration
OPENAI_API_KEY=sk-...
# Anthropic Configuration
ANTHROPIC_API_KEY=sk-ant-...

# === PLATFORM INTEGRATIONS ===
# Discord
DISCORD_API_TOKEN=
DISCORD_APPLICATION_ID=
# Telegram
TELEGRAM_BOT_TOKEN=
# Twitter
TWITTER_API_KEY=
TWITTER_API_SECRET=
TWITTER_ACCESS_TOKEN=
TWITTER_ACCESS_SECRET=

# === DATABASE ===
DATABASE_URL=postgresql://user:pass@db-host:5432/eliza
# Or use PGLite for local development
# DATABASE_URL=pglite://./data/db

# === OPTIONAL SERVICES ===
# Redis (caching)
REDIS_URL=redis://redis-host:6379
# Vector Database (for embeddings)
PINECONE_API_KEY=
PINECONE_ENVIRONMENT=

Step 4: Create Package Configuration

{
  "name": "@eliza/{name}",
  "version": "1.0.0",
  "type": "module",
  "main": "dist/index.js",
  "scripts": {
    "dev": "elizaos dev",
    "start": "elizaos start",
    "test": "elizaos test",
    "build": "tsc",
    "validate": "node scripts/validate-character.js"
  },
  "dependencies": {
    "@elizaos/core": "latest",
    "@elizaos/plugin-bootstrap": "latest",
    "@elizaos/plugin-sql": "latest"
  },
  "devDependencies": {
    "@types/node": "^20.0.0",
    "typescript": "^5.0.0",
    "vitest": "^1.0.0"
  }
}

Step 5: Create Validation Script

// scripts/validate-character.ts

import { validateCharacter } from '@elizaos/core';
import character from '../characters/{name}.js';

const validation = validateCharacter(character);

if (!validation.valid) {
  console.error('❌ Character validation failed:');
  validation.errors.forEach(error => {
    console.error(`  - ${error}`);
  });
  process.exit(1);
}

console.log('✅ Character validation passed');
console.log('\nCharacter Summary:');
console.log(`  Name: ${character.name}`);
console.log(`  Plugins: ${character.plugins?.length || 0}`);
console.log(`  Message Examples: ${character.messageExamples?.length || 0}`);
console.log(`  Knowledge Items: ${character.knowledge?.length || 0}`);

Step 6: Create Tests

// __tests__/character.test.ts

import { describe, it, expect } from 'vitest';
import character from '../characters/{name}';

describe('Character Configuration', () => {
  it('has required fields', () => {
    expect(character.name).toBeDefined();
    expect(character.bio).toBeDefined();
    expect(typeof character.name).toBe('string');
  });

  it('has valid bio format', () => {
    if (Array.isArray(character.bio)) {
      expect(character.bio.length).toBeGreaterThan(0);
      character.bio.forEach(line => {
        expect(typeof line).toBe('string');
        expect(line.length).toBeGreaterThan(0);
      });
    } else {
      expect(typeof character.bio).toBe('string');
      expect(character.bio.length).toBeGreaterThan(0);
    }
  });

  it('has valid message examples', () => {
    expect(character.messageExamples).toBeInstanceOf(Array);
    character.messageExamples?.forEach(conversation => {
      expect(conversation).toBeInstanceOf(Array);
      expect(conversation.length).toBeGreaterThan(0);

      conversation.forEach(message => {
        expect(message).toHaveProperty('name');
        expect(message).toHaveProperty('content');
        expect(message.content).toHaveProperty('text');
      });
    });
  });

  it('has consisten

---

*Content truncated.*