ElizaOS

https://docs.elizaos.ai/
Admin
ElizaOS is a framework for building autonomous AI agents, enabling developers to create...
Tokens:461,625
Snippets:4,725
Trust Score:9.4
Update:2 months ago
Show doc for...
Tokens:
Raw
### Basic start command examples

Source: https://docs.elizaos.ai/cli-reference/start

Common ways to initiate the elizaOS runtime with default or custom settings.

```bash
# Start with default configuration
elizaos start

# Start on custom port
elizaos start --port 8080

# Force reconfiguration
elizaos start --configure
```

--------------------------------

### Advanced configuration examples

Source: https://docs.elizaos.ai/cli-reference/start

Combining reconfiguration, custom ports, and specific character files for production setups.

```bash
# Reconfigure services before starting
elizaos start --configure

# Start with specific character on custom port
elizaos start --character ./my-bot.json --port 4000

# Complete setup for production deployment
elizaos start --character ./production-bot.json --port 3000
```

--------------------------------

### Build and Start Project

Source: https://docs.elizaos.ai/cli-reference/start

Build the project using bun before initiating the start command.

```bash
# Build separately before starting
bun run build
elizaos start
```

--------------------------------

### Run Scenario Examples

Source: https://docs.elizaos.ai/cli-reference/scenario

Examples of running scenarios in standard and live modes.

```bash
# Run a scenario
elizaos scenario run ./tests/greeting.scenario.yaml

# Run in live mode (no mocking)
elizaos scenario run ./tests/api-test.scenario.yaml --live
```

--------------------------------

### Initialize and Start an AI Agent

Source: https://docs.elizaos.ai/

Use these commands to install the CLI, scaffold a new project, and launch the agent.

```bash
bun i -g @elizaos/cli     # Install the CLI
elizaos create            # Create your project
elizaos start             # Your agent is live
```

--------------------------------

### Plugin configuration patterns

Source: https://docs.elizaos.ai/plugin-registry/llm

Example configurations for different plugin setups including primary/fallback strategies and standalone providers.

```json
{
  "plugins": [
    "@elizaos/plugin-anthropic",  // Primary for text
    "@elizaos/plugin-openai"       // Fallback for embeddings
  ]
}
```

```json
{
  "plugins": [
    "@elizaos/plugin-openrouter"  // Handles text and embeddings
  ]
}
```

```json
{
  "plugins": [
    "@elizaos/plugin-openrouter",  // Cloud text & embeddings
    "@elizaos/plugin-ollama"        // Offline fallback
  ]
}
```

--------------------------------

### Build and Start Plugin

Source: https://docs.elizaos.ai/guides/create-a-plugin

Rebuild the plugin and start the ElizaOS runtime from the project root.

```bash
# Build the plugin first
cd plugin-fal-ai
bun run build

# Start from project root  
cd ..
elizaos start
```

--------------------------------

### Build and Start Plugin

Source: https://docs.elizaos.ai/guides/create-a-plugin

Commands to compile the plugin and start the ElizaOS agent to verify the plugin is loaded.

```bash
# Build the plugin first
cd plugin-fal-ai
bun run build

# Go back to project and start
cd ..
elizaos start
```

--------------------------------

### Install Dependencies

Source: https://docs.elizaos.ai/plugins/development

Install the required core elizaOS package and development dependencies.

```bash
# Core dependency
bun add @elizaos/core

# Development dependencies
bun add -d typescript tsup @types/node
```

--------------------------------

### GET /api/system/version

Source: https://docs.elizaos.ai/llms-full.txt

Get the ElizaOS version information.

```APIDOC
## GET /api/system/version

### Description
Get the ElizaOS version information.

### Method
GET

### Endpoint
/api/system/version

### Response
#### Success Response (200)
- **version** (string) - Semantic version string
- **commit** (string) - Git commit hash
- **buildTime** (string) - ISO 8601 build timestamp

#### Response Example
{
  "version": "1.0.0",
  "commit": "abc123def456",
  "buildTime": "2024-12-28T10:00:00Z"
}
```

--------------------------------

### Install local or development plugins

Source: https://docs.elizaos.ai/cli-reference/plugins

Installs a plugin from a local directory or a specific development branch.

```bash
# During development, install from local directory
elizaos plugins add ./path/to/plugin-my-awesome-plugin

# Or install from your development branch
elizaos plugins add my-awesome-plugin --branch feature/new-feature
```

--------------------------------

### System Configuration Response Example

Source: https://docs.elizaos.ai/rest-reference/system/get-system-config

Example JSON response body containing version, environment, features, and limits.

```json
{
  "version": "1.0.0",
  "environment": "production",
  "features": {
    "streaming": true,
    "multiAgent": true,
    "tee": false
  },
  "limits": {
    "maxMessageLength": 10000,
    "maxAgentsPerUser": 10,
    "rateLimitPerMinute": 60
  }
}
```

--------------------------------

### Matrix Scenario Examples

Source: https://docs.elizaos.ai/cli-reference/scenario

Examples of executing matrix configurations with various flags for dry runs, parallelism, filtering, and verbosity.

```bash
# Analyze matrix without executing
elizaos scenario matrix ./matrix-config.yaml --dry-run

# Execute matrix with parallel runs
elizaos scenario matrix ./matrix-config.yaml --parallel 4

# Filter specific combinations
elizaos scenario matrix ./matrix-config.yaml --filter "model=gpt-4"

# Verbose execution
elizaos scenario matrix ./matrix-config.yaml --verbose
```

--------------------------------

### GET /helloworld

Source: https://docs.elizaos.ai/llms-full.txt

An example implementation of a GET route that returns a simple JSON message.

```APIDOC
## GET /helloworld

### Description
A sample endpoint that returns a 'Hello World!' message.

### Method
GET

### Endpoint
/helloworld

### Response
#### Success Response (200)
- **message** (string) - The greeting message.
```

--------------------------------

### Post-creation startup

Source: https://docs.elizaos.ai/cli-reference/create

Navigate to the project directory and start the application.

```bash
cd myproject
elizaos start
# Visit http://localhost:3000
```

--------------------------------

### Get Version Response Example

Source: https://docs.elizaos.ai/rest-reference/system/get-version

Example JSON response body returned by the version endpoint.

```json
{
  "version": "1.0.0",
  "commit": "abc123def456",
  "buildTime": "2024-12-28T10:00:00Z"
}
```

--------------------------------

### Update World Settings

Source: https://docs.elizaos.ai/llms-full.txt

Example of updating world metadata and settings.

```typescript
// Update world settings
await runtime.updateWorld({
  id: worldId,
  name: 'Updated Server Name',
  metadata: {
    settings: { allowBots: true },
    features: ['voice', 'threads']
  }
});
```

--------------------------------

### GET /api/system/config

Source: https://docs.elizaos.ai/llms-full.txt

Retrieves the public system configuration, including version, environment, features, and system limits.

```APIDOC
## GET /api/system/config

### Description
Get the public system configuration. This endpoint returns non-sensitive configuration values that may be useful for clients.

### Method
GET

### Endpoint
/api/system/config

### Response
#### Success Response (200)
- **version** (string) - ElizaOS version
- **environment** (string) - Current environment: development, production
- **features** (object) - Enabled feature flags
- **limits** (object) - System limits and quotas

#### Response Example
{
  "version": "1.0.0",
  "environment": "production",
  "features": {
    "streaming": true,
    "multiAgent": true,
    "tee": false
  },
  "limits": {
    "maxMessageLength": 10000,
    "maxAgentsPerUser": 10,
    "rateLimitPerMinute": 60
  }
}
```

--------------------------------

### Register Discord Message Handler

Source: https://docs.elizaos.ai/llms-full.txt

Implementation example for registering a Discord handler and sending a message.

```typescript
// Register Discord message handler
// Handler signature: (runtime, target, content) => Promise<void>
runtime.registerSendHandler('discord', async (runtime, target, content) => {
  const channel = await discordClient.channels.fetch(target.channelId);
  await channel.send(content.text);
});

// Send message to target
await runtime.sendMessageToTarget(
  { source: 'discord', channelId: '123456', roomId },
  { text: 'Hello from agent!' }
);
```

--------------------------------

### Define an evaluator with training examples

Source: https://docs.elizaos.ai/plugins/components

An example of an evaluator that includes training data to guide the extraction process.

```typescript
const evaluator: Evaluator = {
  name: 'fact-extractor',
  description: 'Extracts facts from conversations',
  examples: [{
    prompt: 'Extract facts from this conversation',
    messages: [
      { name: 'user', content: { text: 'I live in NYC' } },
      { name: 'agent', content: { text: 'NYC is a great city!' } }
    ],
    outcome: 'User lives in New York City'
  }],
  validate: async () => true,
  handler: async (runtime, message, state) => {
    const facts = await extractFacts(state);
    for (const fact of facts) {
      await runtime.factsManager.addFact(fact);
    }
    return facts;
  }
};
```

--------------------------------

### GET /api/agents/{agentId}/runs/{runId}

Source: https://docs.elizaos.ai/llms-full.txt

Get detailed information about a specific agent run, including the full event history.

```APIDOC
## GET /api/agents/{agentId}/runs/{runId}

### Description
Get detailed information about a specific agent run, including the full event history.

### Method
GET

### Endpoint
/api/agents/{agentId}/runs/{runId}

### Parameters
#### Path Parameters
- **agentId** (string) - Required - The UUID of the agent
- **runId** (string) - Required - The UUID of the run

### Response
#### Success Response (200)
- **runId** (string) - Unique run identifier
- **agentId** (string) - Agent that executed this run
- **roomId** (string) - Room where the run occurred
- **startTime** (number) - Unix timestamp when run started
- **endTime** (number) - Unix timestamp when run ended
- **status** (string) - Run status: completed, failed, timeout, cancelled
- **events** (array) - Full event history for this run
- **tokensUsed** (object) - Token usage summary
- **errors** (array) - Array of error messages if any occurred

#### Response Example
{
  "runId": "550e8400-e29b-41d4-a716-446655440001",
  "agentId": "550e8400-e29b-41d4-a716-446655440000",
  "roomId": "660e8400-e29b-41d4-a716-446655440002",
  "startTime": 1703001234567,
  "endTime": 1703001235890,
  "status": "completed",
  "events": [
    {
      "type": "action:started",
      "timestamp": 1703001234600,
      "data": { "action": "REPLY" }
    }
  ],
  "tokensUsed": {
    "promptTokens": 150,
    "completionTokens": 50,
    "totalTokens": 200
  },
  "errors": []
}
```

--------------------------------

### Basic Deployment Examples

Source: https://docs.elizaos.ai/cli-reference/deploy

Common deployment scenarios using default or basic custom settings.

```bash
# Deploy current project with defaults
elizaos deploy

# Deploy with a specific name
elizaos deploy --name my-agent-prod

# Deploy with custom port
elizaos deploy --port 8080
```

--------------------------------

### Define Message Examples for Conversational Behavior

Source: https://docs.elizaos.ai/agents/personality-and-behavior

Use the messageExamples array to provide few-shot examples that guide the agent's tone, problem-solving approach, and knowledge boundaries.

```typescript
messageExamples: [
  // Establishing greeting style
  [
    { name: "{{user}}", content: { text: "Hi" } },
    { name: "Assistant", content: { 
      text: "Hello! 👋 Ready to build something amazing today?" 
    }}
  ],
  
  // Demonstrating problem-solving approach
  [
    { name: "{{user}}", content: { 
      text: "My code isn't working" 
    }},
    { name: "Assistant", content: { 
      text: "I'm here to help! Let's debug this together. Can you tell me:
1. What you expected to happen
2. What actually happened
3. Any error messages you're seeing?" 
    }}
  ],
  
  // Showing knowledge boundaries
  [
    { name: "{{user}}", content: { 
      text: "Can you help with quantum computing?" 
    }},
    { name: "Assistant", content: { 
      text: "While I have basic knowledge of quantum computing concepts, it's not my specialty. I'm most helpful with web development, cloud architecture, and general programming. Would you like me to help with one of those areas instead, or should I share what I do know about quantum computing?" 
    }}
  ],
  
  // Demonstrating teaching style
  [
    { name: "{{user}}", content: { 
      text: "What's a closure?" 
    }},
    { name: "Assistant", content: { 
      text: "Great question! A closure is when a function 'remembers' variables from its outer scope. Here's a simple example:

```javascript
function outer(x) {
  return function inner(y) {
    return x + y; // inner can access x
  }
}

const add5 = outer(5);
console.log(add5(3)); // 8
```

The inner function keeps access to `x` even after `outer` finishes. Think of it like the function packing a 'backpack' of variables it might need later!" 
    }}
  ]
]
```

--------------------------------

### Start an elizaOS project

Source: https://docs.elizaos.ai/cli-reference/overview

Navigate to the project directory and launch the agent.

```bash
# Navigate to your project directory
cd my-agent-project

# Start the project
elizaos start
```

--------------------------------

### Implement HTTP Route

Source: https://docs.elizaos.ai/plugins/architecture

Example implementation of a GET route that returns a JSON response.

```typescript
routes: [
  {
    name: 'hello-world-route',
    path: '/helloworld',
    type: 'GET',
    handler: async (_req: any, res: any) => {
      res.json({ message: 'Hello World!' });
    }
  }
]
```

--------------------------------

### Post-setup initialization

Source: https://docs.elizaos.ai/cli-reference/monorepo

Commands to run immediately after cloning the repository.

```bash
cd eliza                           # Navigate to the cloned directory
bun i && bun run build            # Install dependencies and build
```

--------------------------------

### Implement Run Tracking Lifecycle

Source: https://docs.elizaos.ai/runtime/core

Example of starting, processing, and ending a runtime execution run.

```typescript
// Start tracking a run
const runId = runtime.startRun(roomId);

try {
  // Process messages, actions execute within this run
  await runtime.processActions(message, responses, state);

  // Get results from actions executed in this run
  const results = runtime.getActionResults(message.id);

} finally {
  // Always end the run
  runtime.endRun();
}
```

--------------------------------

### GET /api/messaging/channels/{channelId}

Source: https://docs.elizaos.ai/llms-full.txt

Retrieves information about a specific channel.

```APIDOC
## GET /api/messaging/channels/{channelId}

### Description
Get channel information.

### Method
GET

### Endpoint
/api/messaging/channels/{channelId}

### Parameters
#### Path Parameters
- **channelId** (string) - Required - The ID of the channel.
```

--------------------------------

### Initialize Project with CLI

Source: https://docs.elizaos.ai/llms-full.txt

Basic usage for starting the interactive wizard or using specific command-line options.

```bash
# Interactive mode (recommended)
elizaos create

# With specific options
elizaos create [options] [name]
```

--------------------------------

### GET /api/agents/{agentId}/rooms

Source: https://docs.elizaos.ai/llms-full.txt

Retrieves all rooms associated with a specific agent.

```APIDOC
## GET /api/agents/{agentId}/rooms

### Description
Retrieves all rooms for a specific agent.

### Method
GET

### Endpoint
/api/agents/{agentId}/rooms

### Parameters
#### Path Parameters
- **agentId** (string) - Required - The ID of the agent
```

--------------------------------

### GET /api/agents/{agentId}/runs

Source: https://docs.elizaos.ai/llms-full.txt

List all execution runs for a specific agent. Runs track the lifecycle of message processing including actions executed, evaluators run, and timing data.

```APIDOC
## GET /api/agents/{agentId}/runs

### Description
List all execution runs for a specific agent. Runs track the lifecycle of message processing including actions executed, evaluators run, and timing data.

### Method
GET

### Endpoint
/api/agents/{agentId}/runs

### Parameters
#### Path Parameters
- **agentId** (string) - Required - The UUID of the agent

#### Query Parameters
- **status** (string) - Optional - Filter by run status: `started`, `completed`, `timeout`, `error`, or `all`
- **roomId** (string) - Optional - Filter runs by room ID
- **limit** (number) - Optional - Maximum number of runs to return (max: 100)
- **startedAfter** (number) - Optional - Return only runs started after this Unix timestamp
- **startedBefore** (number) - Optional - Return only runs started before this Unix timestamp

### Headers
- **X-Entity-ID** (string) - Optional - Optional entity ID for Row-Level Security (RLS) filtering

### Response
#### Success Response (200)
- **runs** (array) - Array of run summaries
- **total** (number) - Total number of runs matching the filter
- **hasMore** (boolean) - Whether there are more runs available

#### Response Example
{
  "runs": [
    {
      "runId": "550e8400-e29b-41d4-a716-446655440001",
      "status": "completed",
      "startedAt": 1703001234567,
      "endedAt": 1703001235890,
      "durationMs": 1323,
      "messageId": "770e8400-e29b-41d4-a716-446655440003",
      "roomId": "660e8400-e29b-41d4-a716-446655440002",
      "entityId": "880e8400-e29b-41d4-a716-446655440004",
      "counts": {
        "actions": 2,
        "modelCalls": 3,
        "errors": 0,
        "evaluators": 1
      }
    }
  ],
  "total": 42,
  "hasMore": true
}
```

--------------------------------

### GET /api/messaging/sessions/{sessionId}/messages

Source: https://docs.elizaos.ai/llms-full.txt

Retrieve messages from a conversation session with cursor-based pagination. Messages are returned in reverse chronological order by default.

```APIDOC
## GET /api/messaging/sessions/{sessionId}/messages

### Description
Retrieve messages from a conversation session with cursor-based pagination. Messages are returned in reverse chronological order (newest first) by default.

### Method
GET

### Endpoint
/api/messaging/sessions/{sessionId}/messages

### Parameters
#### Path Parameters
- **sessionId** (string) - Required - The unique identifier of the session

#### Query Parameters
- **limit** (number) - Optional - Number of messages to retrieve (1-100). Default: 50
- **before** (number) - Optional - Timestamp to get messages before (for older messages)
- **after** (number) - Optional - Timestamp to get messages after (for newer messages)

### Response
#### Success Response (200)
- **messages** (array) - Array of messages in the conversation
- **hasMore** (boolean) - Whether more messages are available
- **cursors** (object) - Pagination cursors for fetching additional messages

#### Message Properties
- **id** (string) - Unique message identifier
- **content** (string) - Message content
- **authorId** (string) - UUID of the message author
- **isAgent** (boolean) - Whether the message is from the agent
- **createdAt** (string) - ISO timestamp of message creation
- **metadata** (object) - Message metadata including agent thoughts and actions

#### Cursor Properties
- **before** (number) - Use this timestamp to get older messages
- **after** (number) - Use this timestamp to get newer messages
```

--------------------------------

### Resource Configuration Examples

Source: https://docs.elizaos.ai/cli-reference/deploy

Adjusting CPU, memory, and instance counts for specific workload requirements.

```bash
# Deploy with increased resources
elizaos deploy --cpu 2048 --memory 2048

# Deploy multiple instances for high availability
elizaos deploy --desired-count 3

# Deploy with minimal resources
elizaos deploy --cpu 256 --memory 512
```

--------------------------------

### Activate Plugins in Character Configuration

Source: https://docs.elizaos.ai/cli-reference/plugins

Examples of registering installed plugins in JSON or TypeScript character files to enable them.

```json
{
  "name": "MyAgent",
  "plugins": [
    "@elizaos/plugin-sql",
    "@elizaos/plugin-openai",
    "@elizaos/plugin-discord"
  ],
  "bio": ["Your agent's description"],
  "style": {
    "all": ["conversational", "friendly"]
  }
}
```

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

export const character: Character = {
  name: "MyAgent",
  plugins: [
    // Core plugins
    "@elizaos/plugin-sql",
    
    // Conditional plugins based on environment variables
    ...(process.env.OPENAI_API_KEY ? ["@elizaos/plugin-openai"] : []),
    ...(process.env.DISCORD_API_TOKEN ? ["@elizaos/plugin-discord"] : []),
    ...(process.env.ANTHROPIC_API_KEY ? ["@elizaos/plugin-anthropic"] : [])
  ],
  bio: ["Your agent's description"],
  style: {
    all: ["conversational", "friendly"]
  }
};
```

--------------------------------

### Start Agent

Source: https://docs.elizaos.ai/runtime/services

Command to initialize the agent after configuration.

```bash
bun run start
```

--------------------------------

### Start development mode in project directory

Source: https://docs.elizaos.ai/cli-reference/dev

Standard workflow for navigating to a project and launching the development server.

```bash
# Navigate to your project directory
cd my-agent-project

# Start development mode
elizaos dev
```

--------------------------------

### Basic CLI Usage

Source: https://docs.elizaos.ai/cli-reference/create

Run the interactive wizard or view help documentation for the create command.

```bash
# Interactive mode (recommended)
elizaos create

# With specific options
elizaos create [options] [name]
```

```bash
# View detailed help
elizaos create --help
```

--------------------------------

### Get Job Response Example

Source: https://docs.elizaos.ai/rest-reference/jobs/get-job

A sample JSON response showing a successfully completed job with its associated result data.

```json
{
  "jobId": "job_550e8400-e29b-41d4-a716-446655440001",
  "status": "completed",
  "createdAt": 1703001234567,
  "startedAt": 1703001234600,
  "completedAt": 1703001235890,
  "result": {
    "text": "Here is the processed response.",
    "actions": ["REPLY"]
  }
}
```

--------------------------------

### OpenAPI specification for Get room memories

Source: https://docs.elizaos.ai/rest-reference/memory/get-room-memories

Defines the endpoint parameters including agentId, roomId, count, unique, start, and end filters.

```yaml
openapi: 3.1.0
info:
  title: Eliza OS API
  description: >-
    API documentation for Eliza OS v1.2.0 - A flexible and scalable AI agent
    framework.


    This API is designed to be used with a locally running Eliza instance.
    Endpoints allow for creating,

    managing, and interacting with AI agents through a REST interface.


    The API is organized into the following domains:

    - **System**: System-wide operations and environment management

    - **Agents**: Agent lifecycle and management operations

    - **Memory**: Agent memory and room management

    - **Messaging**: Message handling, channels, and servers

    - **Audio**: Audio processing and speech synthesis

    - **Media**: File upload and media management

    - **TEE**: Trusted Execution Environment operations

    - **WebSocket**: Real-time communication via Socket.IO
  version: 1.2.0
  contact:
    name: Eliza OS Community
    url: https://github.com/elizaos/eliza
servers:
  - url: http://localhost:3000
    description: Local development server
security: []
tags:
  - name: system
    description: System-wide operations and environment management
  - name: agents
    description: Operations for managing AI agents
  - name: memory
    description: Operations for managing agent memories
  - name: rooms
    description: Operations for managing rooms
  - name: messaging
    description: Operations for messages, channels, and servers
  - name: audio
    description: Operations for speech and audio processing
  - name: media
    description: Operations for file uploads and media management
  - name: logs
    description: Operations for accessing system and agent logs
  - name: tee
    description: Trusted Execution Environment operations
  - name: websocket
    description: Real-time WebSocket communication
paths:
  /api/agents/{agentId}/rooms/{roomId}/memories:
    get:
      tags:
        - memory
        - rooms
        - agents
      summary: Get room memories
      description: Retrieves memories for a specific room
      operationId: getRoomMemories
      parameters:
        - name: agentId
          in: path
          required: true
          schema:
            type: string
            format: uuid
          description: ID of the agent
        - name: roomId
          in: path
          required: true
          schema:
            type: string
            format: uuid
          description: ID of the room
        - name: count
          in: query
          schema:
            type: integer
            default: 50
          description: Number of memories to retrieve
        - name: unique
          in: query
          schema:
            type: boolean
            default: true
          description: Return only unique memories
        - name: start
          in: query
          schema:
            type: integer
          description: Start timestamp filter
        - name: end
          in: query
          schema:
            type: integer
          description: End timestamp filter
      responses:
        '200':
          description: Room memories retrieved successfully
          content:
            application/json:
              schema:
                type: object
                properties:
                  success:
                    type: boolean
                    example: true
                  data:
                    type: array
                    items:
                      $ref: '#/components/schemas/Memory'
        '400':
          description: Invalid agent ID or room ID
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '404':
          description: Agent or room not found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '500':
          description: Error retrieving memories
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
components:
  schemas:
    Memory:
      type: object
      properties:
        id:
          type: string
          format: uuid
          description: Unique identifier for the memory
        entityId:
          type: string
          format: uuid
          description: ID of the entity associated with this memory
        agentId:
          type: string
          format: uuid
          description: ID of the agent associated with this memory
        roomId:
          type: string
          format: uuid
          description: ID of the room this memory belongs to
        createdAt:
          type: integer
          format: int64
```

--------------------------------

### Plugin Creation and Development

Source: https://docs.elizaos.ai/cli-reference/publish

Initial steps to scaffold a new plugin and begin development.

```bash
# Create a new plugin from the template
elizaos create -t plugin my-awesome-plugin
cd my-awesome-plugin

# Install dependencies and start development
bun install
elizaos dev
```

--------------------------------

### Troubleshoot dependency installation

Source: https://docs.elizaos.ai/cli-reference/create

Manually install dependencies or clear the Bun cache if automatic installation fails.

```bash
# If bun install fails, try manual installation
cd myproject
bun install

# For network issues, clear cache and retry
bun pm cache rm
bun install
```

--------------------------------

### Install Bun in Git Bash

Source: https://docs.elizaos.ai/installation

Installs Bun using the PowerShell installer command within Git Bash.

```bash
powershell -c "irm bun.sh/install.ps1 | iex"
```

--------------------------------

### Basic usage of elizaos start

Source: https://docs.elizaos.ai/cli-reference/start

The standard command syntax for launching elizaOS.

```bash
elizaos start [options]
```

--------------------------------

### Initialize elizaos project

Source: https://docs.elizaos.ai/guides/deploy-a-project

Create a new project directory and initialize the agent structure.

```bash
elizaos create my-agent --type project
cd my-agent
```

--------------------------------

### Install or Verify CLI

Source: https://docs.elizaos.ai/cli-reference/update

Commands to check for global installation and install the elizaos CLI, including permission handling for macOS/Linux.

```bash
# Check if the CLI is installed globally
bun pm ls -g @elizaos/cli

# If not, install it
bun install -g @elizaos/cli

# On macOS/Linux, you may need sudo
sudo bun install -g @elizaos/cli
# Or fix permissions on your bun directory
sudo chown -R $(whoami) ~/.bun
```

--------------------------------

### Initialize the Project

Source: https://docs.elizaos.ai/plugins/development

Commands to create a new directory and initialize a Bun project.

```bash
mkdir plugin-my-custom
cd plugin-my-custom
bun init
```

--------------------------------

### Basic Usage

Source: https://docs.elizaos.ai/cli-reference/deploy

The standard command syntax for initiating a deployment.

```bash
elizaos deploy [options]
```

--------------------------------

### Clean Install and Version Conflict Resolution

Source: https://docs.elizaos.ai/installation

Removes existing CLI installations across multiple package managers to ensure a clean state for a fresh Bun-based installation.

```bash
# Remove from all package managers
bun remove -g @elizaos/cli
npm uninstall -g @elizaos/cli
pnpm remove -g @elizaos/cli

# Verify removal
which elizaos
# Should return nothing or "not found"

# Fresh install with bun only
bun i -g @elizaos/cli

# Verify correct version
elizaos --version
```

--------------------------------

### Swap Intent Example

Source: https://docs.elizaos.ai/plugin-registry/defi/evm/defi-operations-flow

Example of a user intent for a token swap.

```text
User: Swap 1 ETH for USDC
```

--------------------------------

### Transfer Intent Example

Source: https://docs.elizaos.ai/plugin-registry/defi/evm/defi-operations-flow

Example of a user intent for a token transfer.

```text
User: Send 0.1 ETH to alice.eth
```

--------------------------------

### Start Agent Runtime

Source: https://docs.elizaos.ai/cli-reference/agent

Launches the agent runtime server.

```bash
# Start the agent runtime server
elizaos start
```

--------------------------------

### Configure Development Setup

Source: https://docs.elizaos.ai/projects/environment-variables

Sets the environment to development with default behaviors for convenience.

```bash
NODE_ENV=development
# ELIZA_SERVER_AUTH_TOKEN=  # Unset for easy development
# ELIZA_UI_ENABLE=         # Unset for automatic behavior (UI enabled)
```

--------------------------------

### elizaos agent start

Source: https://docs.elizaos.ai/llms-full.txt

Starts an agent using a character profile.

```APIDOC
## CLI Command: elizaos agent start

### Description
Starts an agent with a character profile. Can be used with alias 's'.

### Required Options
- **-n, --name** (string) - Required - Name of an existing agent to start.
- **--path** (string) - Required - Path to local character JSON file.
- **--remote-character** (string) - Required - URL to remote character JSON file.

### Additional Options
- **-r, --remote-url <url>** (string) - Optional - URL of the remote agent runtime.
- **-p, --port <port>** (number) - Optional - Port to listen on.
- **--auth-token <token>** (string) - Optional - API authentication token.
```

--------------------------------

### Troubleshoot installation failures

Source: https://docs.elizaos.ai/cli-reference/plugins

Clears the local cache to resolve installation issues.

```bash
# Clear cache and retry
rm -rf ~/.eliza/cache
elizaos plugins add plugin-name
```

--------------------------------

### Prepare action directory

Source: https://docs.elizaos.ai/guides/create-a-plugin

Create the actions directory and copy the plugin template to start building the new action.

```bash
mkdir src/actions
cp src/plugin.ts src/actions/generateVideo.ts
```

--------------------------------

### Governance Flow Examples

Source: https://docs.elizaos.ai/llms-full.txt

Examples of user intents for proposal creation and voting.

```text
User: Create a proposal to increase treasury allocation
→ Plugin creates proposal transaction with targets, values, and description
```

```text
User: Vote FOR on proposal 42
→ Plugin casts vote with correct proposal ID and support value
```

--------------------------------

### Bridge Intent Example

Source: https://docs.elizaos.ai/plugin-registry/defi/evm/defi-operations-flow

Example of a user intent for bridging tokens between chains.

```text
User: Bridge 100 USDC from Ethereum to Base
```

--------------------------------

### Launch agents via CLI

Source: https://docs.elizaos.ai/guides/add-multiple-agents

Start the project to initialize all configured agents simultaneously.

```bash
elizaos start
```

--------------------------------

### Transfer Response Example

Source: https://docs.elizaos.ai/plugin-registry/defi/evm/defi-operations-flow

Example of the agent's response after a successful transfer.

```text
Agent: Successfully transferred 0.1 ETH to alice.eth
Transaction: https://etherscan.io/tx/[hash]
```

--------------------------------

### Create a new elizaOS project

Source: https://docs.elizaos.ai/cli-reference/overview

Initialize a new project using the interactive wizard or by providing a project name directly.

```bash
# Create a new project using the interactive wizard
elizaos create

# Or specify a name directly
elizaos create my-agent-project
```
Tokens:
Raw