### Local Development Setup

Source: https://lobehub.com/docs/development/agents/agent-index-and-submit

Clone the repository and install dependencies for local development. This is the initial step for contributing or running the project locally.

```bash
git clone https://github.com/lobehub/lobe-chat-agents.git
cd lobe-chat-agents
bun install
```

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

### Successful Startup Log Example (Local Mode)

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

This log output indicates that the LobeHub server has started successfully and is accessible locally.

```log
[Database] Start to migration...
✅ database migration pass.
-------------------------------------
  ▲ Next.js 16.x.x
  - Local:        http://localhost:3210
  - Network:      http://0.0.0.0:3210

 ✓ Starting...
 ✓ Ready in 95ms
```

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

### Successful Startup Log Example (Port Mode)

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

This log output confirms that the LobeHub server has started successfully and is accessible via the network IP.

```log
[Database] Start to migration...
✅ database migration pass.
-------------------------------------
  ▲ Next.js 16.x.x
  - Local:        http://your_server_ip:3210
  - Network:      http://0.0.0.0:3210
 ✓ Starting...
 ✓ Ready in 95ms
```

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

### Production Redis Configuration with Authentication

Source: https://lobehub.com/docs/self-hosting/advanced/redis

Example configuration for a production Redis setup, including authentication, TLS, and a standard prefix.

```shell
REDIS_URL=redis://localhost:6379
REDIS_PASSWORD=your-strong-password
REDIS_PREFIX=lobechat
REDIS_TLS=true
```

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

### Example Settings Object

Source: https://lobehub.com/docs/self-hosting/advanced/settings-url-share

An example of a settings object in JSON format that can be used for importing.

```json
{
  "keyVaults": {
    "openai": {
      "apiKey": "user-key"
    }
  }
}
```

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

### Install Dokploy

Source: https://lobehub.com/docs/self-hosting/platform/dokploy

Use this command to install Dokploy on your server.

```Shell
curl -sSL https://dokploy.com/install.sh | sh
```

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

### System Prompt for Customer Support Agent

Source: https://lobehub.com/docs/usage/getting-started/resource

This is an example of a system prompt to guide an agent's behavior when accessing product documentation. It emphasizes searching the knowledge base, citing sources, and handling uncertainty.

```plaintext
You are a customer support agent with access to our product documentation.

When answering product questions:
1. Always search the knowledge base first
2. Cite specific documents in your responses
3. If unsure, read the full document for context
4. If you can't find info, let the user know
```

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

### Install and Run Codex CLI

Source: https://lobehub.com/docs/usage/agent/codex

This snippet shows how to install the Codex CLI and run it once to initiate the sign-in process. Sign-in is handled at the CLI level.

```bash
npm install -g @lobehub/codex
codex
```

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

### Best Practice: Start with -all

Source: https://lobehub.com/docs/self-hosting/advanced/model-list

Recommends starting model list configuration with `-all` to hide default models, followed by explicitly adding desired models like `gpt-4o` and `gpt-4o-mini`.

```shell
OPENAI_MODEL_LIST=-all,+gpt-4o,+gpt-4o-mini
```

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

### Install Project Dependencies with PNPM

Source: https://lobehub.com/docs/development/basic/setup-development

Install all necessary project dependencies using PNPM.

```shell
pnpm i
```

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

### Example Model List Configuration

Source: https://lobehub.com/docs/self-hosting/advanced/model-list

Illustrates adding specific models, removing others, and customizing display names. This example shows how to add `qwen-7b-chat` and `glm-6b`, remove `gpt-3.5-turbo`, and rename `gpt-4-0125-preview` to `gpt-4-turbo`.

```plaintext
+qwen-7b-chat,+glm-6b,-gpt-3.5-turbo,gpt-4-0125-preview=gpt-4-turbo
```

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

### Example Prompts for Video Generation

Source: https://lobehub.com/docs/usage/getting-started/generation

These examples demonstrate how to describe scenes, motion, and camera techniques for video generation. Focus on verbs and shot language for better results.

```plaintext
"A red fox trotting through fresh snow at golden hour, breath visible in the cold air, slow tracking shot, cinematic"
```

```plaintext
"An astronaut floating into a colorful nebula, slow dolly-in, dreamy atmosphere, soft volumetric light"
```

```plaintext
"A cup of coffee being poured in macro slow motion, steam rising, shallow depth of field, commercial product shot"
```

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

### PostgreSQL Database Connection URL Example

Source: https://lobehub.com/docs/self-hosting/platform/dokploy

This is an example of a PostgreSQL database connection URL. Ensure the port is unused and accessible externally.

```Shell
postgresql://postgres:wAbLxfXSwkxxxxxx@45.577.281.48:5432/postgres
```

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

### Examples of Model Extension Configurations

Source: https://lobehub.com/docs/self-hosting/advanced/model-list

Provides specific examples of how to configure models with their maximum token limits and various extension capabilities such as Function Calling, Vision, and Image Generation.

```plaintext
chatglm-6b=ChatGLM 6B<4096>
```

```plaintext
spark-v3.5=讯飞星火 v3.5<8192:fc>
```

```plaintext
gemini-1.5-flash-latest=Gemini 1.5 Flash<16000:vision>
```

```plaintext
o3-mini=OpenAI o3-mini<200000:reasoning:fc>
```

```plaintext
qwen-max-latest=Qwen Max<32768:search:fc>
```

```plaintext
gpt-4-all=ChatGPT Plus<128000:fc:vision:file>
```

```plaintext
gemini-2.0-flash-exp-image-generation=Gemini 2.0 Flash (Image Generation) Experimental<32768:imageOutput:vision>
```

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

### HTTP/HTTPS Connection Example

Source: https://lobehub.com/docs/usage/community/mcp-market

Example configuration for an MCP server using HTTP/HTTPS, suitable for cloud services and remote APIs.

```JSON
{
  "type": "http",
  "url": "https://api.example.com/mcp",
  "headers": {
    "Authorization": "Bearer YOUR_API_TOKEN"
  }
}
```

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

### Effective Prompt Structure Example

Source: https://lobehub.com/docs/usage/getting-started/generation

This example demonstrates a structured approach to writing effective prompts for image generation, outlining key components like subject, style, setting, and technical details.

```plaintext
[Subject] [Style/Medium] [Setting/Background] [Lighting] [Mood] [Technical details]
```

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

### View Usage Statistics

Source: https://lobehub.com/docs/usage/subscription/how-to-check-credit-usage

This example describes how to view the current month's computing usage on the Usage Statistics page.

```text
On the **Usage Statistics** page, you can view the computing usage for the current month.
```

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

### STDIO Connection Example

Source: https://lobehub.com/docs/usage/community/mcp-market

Example configuration for an MCP server using the STDIO connection type, typically for local tools and scripts.

```JSON
{
  "type": "stdio",
  "command": "npx",
  "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/workspace"]
}
```

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

### Start LobeHub Development Server

Source: https://lobehub.com/docs/development/basic/setup-development

Launch the LobeHub development server to access the application in your browser.

```shell
bun run dev
```

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

### S3 Bucket Name Example

Source: https://lobehub.com/docs/self-hosting/environment-variables/s3

Enter the name of your S3 storage bucket. This is a required field.

```bash
my-bucket
```

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

### ComfyUI Authentication Configuration Examples

Source: https://lobehub.com/docs/development/basic/comfyui-development

Provides examples for configuring LobeComfyUI with different authentication methods: none, basic, bearer token, and custom headers. Ensure the baseURL is correctly set for each configuration.

```typescript
// No authentication
const comfyUI = new LobeComfyUI({
  baseURL: 'http://localhost:8000',
  authType: 'none'
});
```

```typescript
// Basic authentication
const comfyUI = new LobeComfyUI({
  baseURL: 'https://your-comfyui-server.com',
  authType: 'basic',
  username: 'your-username',
  password: 'your-password'
});
```

```typescript
// Bearer Token
const comfyUI = new LobeComfyUI({
  baseURL: 'https://your-comfyui-server.com',
  authType: 'bearer',
  apiKey: 'your-api-key'
});
```

```typescript
// Custom headers
const comfyUI = new LobeComfyUI({
  baseURL: 'https://your-comfyui-server.com',
  authType: 'custom',
  customHeaders: {
    'X-API-Key': 'your-custom-key',
    'Authorization': 'Custom your-token'
  }
});
```

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

### Image Generation Prompt Examples

Source: https://lobehub.com/docs/usage/getting-started/generation

Illustrative examples of detailed prompts for image generation, showcasing how to specify subject, style, lighting, and quality for desired outcomes.

```plaintext
"A futuristic city skyline at sunset, digital art, cyberpunk style, neon lights reflecting on wet streets, cinematic lighting, 4K detail"
```

```plaintext
"A cozy coffee shop interior, watercolor illustration, warm golden light streaming through windows, potted plants on windowsills, soft and inviting atmosphere"
```

```plaintext
"A product photo of a minimalist leather wallet on a clean white background, studio lighting, sharp focus, commercial photography style"
```

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

### Create Deployment Directory

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

Execute these commands to create a directory for configuration and database files and navigate into it.

```bash
mkdir lobehub && cd lobehub
```

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

### Generate JWT Client Secret with Node.js

Source: https://lobehub.com/docs/self-hosting/auth/providers/apple

Use this Node.js example to generate a JWT client secret required for Apple Sign In. Ensure you have the 'jsonwebtoken' package installed. Replace placeholders with your actual Team ID, Services ID, and Key ID.

```javascript
// Example using Node.js
const jwt = require('jsonwebtoken');
const fs = require('fs');

const privateKey = fs.readFileSync('AuthKey_XXXXX.p8');
const token = jwt.sign({}, privateKey, {
  algorithm: 'ES256',
  expiresIn: '180d', // Max 6 months
  issuer: 'YOUR_TEAM_ID',
  audience: 'https://appleid.apple.com',
  subject: 'YOUR_SERVICES_ID', // Client ID
  keyid: 'YOUR_KEY_ID',
});

```

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

### Install Claude Code CLI with Script

Source: https://lobehub.com/docs/usage/agent/claude-code

Use this command to install the Claude Code CLI using the recommended installation script. Ensure you have curl installed.

```shell
curl -fsSL https://claude.ai/install.sh | bash
```

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

### Install Ollama on Linux

Source: https://lobehub.com/docs/usage/providers/ollama

Installs Ollama on Linux using a curl script. Alternatively, manual installation is available.

```Shell
curl -fsSL https://ollama.com/install.sh | sh
```

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

### Start Docker Services

Source: https://lobehub.com/docs/development/basic/setup-development

Launch all required development services, including PostgreSQL, Redis, RustFS, and SearXNG, using Docker Compose.

```shell
bun run dev:docker
```

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

### Install vLLM using Conda

Source: https://lobehub.com/docs/usage/providers/vllm

Installs vLLM within a Conda-managed Python environment. Ensure Conda is installed and configured.

```Shell
conda create -n myenv python=3.12 -y
conda activate myenv
pip install vllm
```

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

### Install Claude Code CLI with Homebrew

Source: https://lobehub.com/docs/usage/agent/claude-code

Install the Claude Code CLI on macOS using Homebrew. This is an alternative installation method.

```shell
brew install --cask claude-code
```

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

### Path-Style vs Virtual-Host URL Examples

Source: https://lobehub.com/docs/self-hosting/environment-variables/s3

Illustrates the URL structure differences between path-style and virtual-host access modes for S3 buckets and objects.

```text
path-style: s3.example.net/mybucket/config.env
virtual-host: mybucket.s3.example.net/config.env
```

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

### Install vLLM using uv

Source: https://lobehub.com/docs/usage/providers/vllm

Installs vLLM within a new Python environment managed by uv. Ensure uv is installed first.

```Shell
uv venv myenv --python 3.12 --seed
source myenv/bin/activate
uv pip install vllm
```

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

### Install Codex CLI with Homebrew

Source: https://lobehub.com/docs/usage/agent/codex

Use this command to install the Codex CLI on macOS using Homebrew. This is a recommended installation method for macOS users.

```bash
brew install --cask codex
```

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

### Configure SSRF Protection Scenarios

Source: https://lobehub.com/docs/self-hosting/environment-variables/basic

Examples demonstrating SSRF protection configurations for different deployment scenarios. Scenario 1 is recommended for public deployments, blocking all private IPs. Scenario 2 allows private IPs for internal deployments. Scenario 3 is a hybrid approach.

```bash
# Scenario 1: Public deployment (recommended)
# Block all private IP addresses for security
SSRF_ALLOW_PRIVATE_IP_ADDRESS=0
```

```bash
# Scenario 2: Internal deployment
# Allow all private IP addresses to access internal image servers
SSRF_ALLOW_PRIVATE_IP_ADDRESS=1
```

```bash
# Scenario 3: Hybrid deployment (most common)
# Block private IPs by default, but allow specific trusted internal servers
SSRF_ALLOW_PRIVATE_IP_ADDRESS=0
SSRF_ALLOW_IP_ADDRESS_LIST=192.168.1.100,10.0.0.50
```

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

### Install Codex CLI with npm

Source: https://lobehub.com/docs/usage/agent/codex

Use this command to install the Codex CLI globally via npm. Ensure Node.js and npm are installed on your system.

```bash
npm install -g @openai/codex
```

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

### Configure Migration Environment Variables

Source: https://lobehub.com/docs/self-hosting/migration/v2/auth/clerk-to-betterauth

Set up the .env file with necessary variables for migration mode, database connections, Clerk API keys, and optional settings like database driver and dry run mode.

```shell
# ============================================ 
# Migration mode: test or prod
# Recommended: Start with test mode to verify on a test database,
# then switch to prod after confirming everything works
# ============================================ 
CLERK_TO_BETTERAUTH_MODE=test

# ============================================ 
# Database connection (uses corresponding variable based on mode)
# TEST_ prefix for test environment, PROD_ prefix for production
# ============================================ 
TEST_CLERK_TO_BETTERAUTH_DATABASE_URL=postgresql://user:pass@test-host:5432/testdb
PROD_CLERK_TO_BETTERAUTH_DATABASE_URL=postgresql://user:pass@prod-host:5432/proddb

# ============================================ 
# Clerk API keys (for exporting user data via API)
# Get from Clerk Dashboard: Configure → Developers → API Keys
# ============================================ 
TEST_CLERK_TO_BETTERAUTH_CLERK_SECRET_KEY=sk_test_xxx
PROD_CLERK_TO_BETTERAUTH_CLERK_SECRET_KEY=sk_live_xxx

# ============================================ 
# Database driver (optional)
# neon: Neon serverless driver (default)
# node: node-postgres driver
# ============================================ 
CLERK_TO_BETTERAUTH_DATABASE_DRIVER=neon

# ============================================ 
# Dry Run mode (optional)
# Set to 1 to only print logs without modifying the database
# Recommended: Enable for first run, disable after verification
# ============================================ 
CLERK_TO_BETTERAUTH_DRY_RUN=1
```

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

### Install Comfy-Manager Manually

Source: https://lobehub.com/docs/usage/providers/comfyui

Use this method for a direct installation of Comfy-Manager by cloning the repository into your ComfyUI custom_nodes directory. Ensure you restart the ComfyUI server after installation.

```shell
# Navigate to ComfyUI's custom_nodes directory
cd ComfyUI/custom_nodes

# Clone the Comfy-Manager repository
git clone https://github.com/ltdrdata/ComfyUI-Manager.git

# Restart the ComfyUI server
# After restarting, you should see a "Manager" button in the UI
```

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

### Read Documentation with Agent

Source: https://lobehub.com/docs/usage/help

Use this prompt to instruct an agent to read the help documentation.

```shell
Read https://lobehub.com/docs/usage/help
```

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

### Install Comfy-Manager with One-Click Script

Source: https://lobehub.com/docs/usage/providers/comfyui

This script automates the Comfy-Manager installation process. Run it from your ComfyUI root directory.

```shell
# Run from the ComfyUI root directory
curl -fsSL https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/install.sh | bash
```

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

### Example S3_ENDPOINT Configuration

Source: https://lobehub.com/docs/self-hosting/advanced/s3

Example of how to set the S3_ENDPOINT for Cloudflare R2. Ensure the endpoint does not include the bucket name.

```shell
S3_ENDPOINT=https://0b33a03b5c993fd2f453379dc36558e5.r2.cloudflarestorage.com
```

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

### Download Docker Compose and Environment Files

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

Download the docker-compose.yml and .env.example files to set up your LobeHub deployment. Rename .env.example to .env to apply your custom configurations.

```Shell
curl -O https://raw.githubusercontent.com/lobehub/lobehub/HEAD/docker-compose/deploy/docker-compose.yml
curl -O https://raw.githubusercontent.com/lobehub/lobehub/HEAD/docker-compose/deploy/.env.example
mv .env.example .env
```

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

### Start Docker Compose Services

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

Use this command to start all services defined in the docker-compose.yml file in detached mode.

```bash
docker compose up -d
```

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

### Install Ollama Model via CLI

Source: https://lobehub.com/docs/usage/providers/ollama

Installs a specific Ollama model, such as llama3, using the Ollama command-line interface.

```Shell
ollama pull llama3
```

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

### Fetch and Execute Deployment Script

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

Download and run the LobeHub setup script. The '-l en' flag sets the language to English.

```bash
bash <(curl -fsSL https://lobe.li/setup.sh) -l en
```

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

### Install Claude Code CLI

Source: https://lobehub.com/docs/usage/agent/claude-code

Install the Claude Code CLI using npm. This is a prerequisite for using the agent in LobeHub.

```bash
npm install -g claude

```

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

### Local Development Redis Configuration

Source: https://lobehub.com/docs/self-hosting/advanced/redis

Example configuration for setting up Redis for local development, including URL and a development-specific prefix.

```shell
REDIS_URL=redis://localhost:6379
REDIS_PREFIX=lobechat-dev
```

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

### Example Task Prompt

Source: https://lobehub.com/docs/usage/agent/scheduled-task

This is an example of a prompt that can be used for a scheduled task. It specifies the content to analyze and the desired format for the output.

```plaintext
Analyze today's top tech news and summarize:
1. Major product launches
2. Funding announcements
3. Industry trends
Format as a brief executive summary.
```

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

### Configuring Multiple Model Providers

Source: https://lobehub.com/docs/self-hosting/advanced/model-list

Demonstrates how to configure model lists for multiple providers simultaneously, including OpenAI, Anthropic, and Google. Each provider has its API key and model list defined separately.

```shell
# OpenAI — curated list
OPENAI_API_KEY=sk-...
OPENAI_MODEL_LIST=-all,+gpt-4o,+gpt-4o-mini

# Anthropic — long context backup
ANTHROPIC_API_KEY=sk-ant-...
ANTHROPIC_MODEL_LIST="+claude-3-opus-20240229=Claude 3 Opus<200000:vision:fc>,+claude-3-5-sonnet-20241022=Claude 3.5 Sonnet<200000:vision:fc"

# Google
GOOGLE_API_KEY=...
GOOGLE_MODEL_LIST="+gemini-1.5-pro=Gemini 1.5 Pro<1000000:vision:fc>"
```

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

### Verify Comfy-Manager Installation

Source: https://lobehub.com/docs/usage/providers/comfyui

After installing Comfy-Manager, verify its successful integration by restarting the ComfyUI server and checking for the 'Manager' button in the UI.

```text
1. Restart the ComfyUI server
2. Visit http://localhost:8000
3. You should see a "Manager" button in the bottom-right corner
```

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

### Inspect RustFS Service and Initialization Logs

Source: https://lobehub.com/docs/self-hosting/platform/docker-compose

View logs for the RustFS service and its initialization process to troubleshoot file upload failures. Also, verify if the required bucket has been created.

```shell
docker compose logs rustfs
docker compose logs rustfs-init
```

```shell
docker compose exec rustfs ls /data/lobe
```

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

### Start vLLM server

Source: https://lobehub.com/docs/usage/providers/vllm

Starts a vLLM server with the specified model. The server is compatible with the OpenAI API protocol and defaults to http://localhost:8000.

```Shell
vllm serve Qwen/Qwen2.5-1.5B-Instruct
```

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

### Best Practice: Use Descriptive Display Names

Source: https://lobehub.com/docs/self-hosting/advanced/model-list

Illustrates using user-friendly and meaningful display names for models, such as 'GPT-4o (Recommended)' and 'GPT-4o Mini (Fast & Cheap)', to improve user experience.

```shell
OPENAI_MODEL_LIST="gpt-4o=GPT-4o (Recommended),gpt-4o-mini=GPT-4o Mini (Fast & Cheap)"
```

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

### Configure Development Environment Variables

Source: https://lobehub.com/docs/development/basic/setup-development

Copy example environment files for Docker services and the Next.js development server. These files should be edited to match your specific configuration needs.

```shell
# Docker services configuration
cp docker-compose/dev/.env.example docker-compose/dev/.env

# Next.js development server configuration
cp .env.example.development .env
```