Pixabay MCP Server
# Pixabay MCP Server
[English](#english) | [中文](#中文)
---
## English
A Model Context Protocol (MCP) server that enables AI assistants to search for images and videos on [Pixabay](https://pixabay.com).
### Features
- 🖼️ **search_images** - Search for photos, illustrations, and vectors
- 🎬 **search_videos** - Search for videos and animations
### Installation
#### Method 1: Quick Start with uvx (Recommended)
The easiest way to use this MCP server is with `uvx`. No manual cloning required!
1. Get your [Pixabay API Key](https://pixabay.com/api/docs/)
2. Add the following to your MCP client configuration:
```json
{
"mcpServers": {
"pixabay": {
"command": "uvx",
"args": [
"https://github.com/helloHupc/pixabay_mcp.git"
],
"env": {
"PIXABAY_API_KEY": "your-api-key-here"
}
}
}
}
```
3. Restart your MCP client and start using!
#### Method 2: Local Development
For development or to customize the code, clone the repository locally:
```bash
git clone https://github.com/helloHupc/pixabay_mcp.git
cd pixabay_mcp
```
Then configure your MCP client:
```json
{
"mcpServers": {
"pixabay": {
"command": "uv",
"args": [
"run",
"--directory", "/path/to/pixabay_mcp",
"python", "src/pixabay_mcp/server.py"
],
"env": {
"PIXABAY_API_KEY": "your-api-key-here"
}
}
}
}
```
Make sure to replace `/path/to/pixabay_mcp` with your actual local path.
### Get Your API Key
1. Create a free account at [Pixabay](https://pixabay.com/accounts/register/)
2. Go to [API Documentation](https://pixabay.com/api/docs/) and copy your API key
### Quick Start
1. **Get your API key from Pixabay**
2. **Copy** configuration from Method 1 above
3. **Replace `your-api-key-here`** with your actual API key
4. **Add to your MCP client settings**
5. **Restart your MCP client**
6. **Start searching!**
### Usage Examples
Once configured, you can ask your AI assistant:
- "Search for photos of yellow flowers"
- "Find some nature videos"
- "Look for vector illustrations of cats"
### Project Structure
```
pixabay_mcp/
├── src/
│ └── pixabay_mcp/
│ ├── __init__.py
│ └── server.py # Main MCP server implementation
├── pyproject.toml # Project configuration
├── uv.lock # Dependency lock file
├── README.md # This file
├── LICENSE # MIT License
└── .gitignore # Git ignore rules
```
### License
MIT License
---
## 中文
一个 MCP (Model Context Protocol) 服务,让 AI 助手能够在 [Pixabay](https://pixabay.com) 上搜索图片和视频。
### 功能
- 🖼️ **search_images** - 搜索照片、插画和矢量图
- 🎬 **search_videos** - 搜索视频和动画
### 安装
#### 方法 1:使用 uvx 快速开始(推荐)
最简单的使用方式,使用 `uvx` 直接从 Gitee 运行,无需手动克隆!
1. 获取你的 [Pixabay API 密钥](https://pixabay.com/api/docs/)
2. 在 MCP 客户端配置中添加以下内容:
```json
{
"mcpServers": {
"pixabay": {
"command": "uvx",
"args": [
"https://github.com/helloHupc/pixabay_mcp.git"
],
"env": {
"PIXABAY_API_KEY": "你的API密钥"
}
}
}
}
```
3. 重启 MCP 客户端,开始使用!
#### 方法 2:本地开发调试
用于开发或自定义代码,将仓库克隆到本地:
```bash
git clone https://github.com/helloHupc/pixabay_mcp.git
cd
```
然后配置 MCP 客户端:
```json
{
"mcpServers": {
"pixabay": {
"command": "uv",
"args": [
"run",
"--directory", "/path/to/pixabay_mcp",
"python", "src/pixabay_mcp/server.py"
],
"env": {
"PIXABAY_API_KEY": "你的API密钥"
}
}
}
}
```
请将 `/path/to/pixabay_mcp` 替换为你的实际本地路径。
### 获取 API 密钥
1. 在 [Pixabay](https://pixabay.com/accounts/register/) 注册免费账号
2. 访问 [API 文档页面](https://pixabay.com/api/docs/) 复制你的 API 密钥
### 快速开始
1. **从 Pixabay 获取你的 API 密钥**
2. **复制上面方法 1 中的配置**
3. **将 `你的API密钥`** 替换为你的实际 API 密钥
4. **添加到你的 MCP 客户端设置**
5. **重启 MCP 客户端**
6. **开始搜索!**
### 使用示例
配置完成后,你可以这样问 AI 助手:
- "帮我搜索黄色花朵的图片"
- "找一些自然风景的视频"
- "搜索猫咪的矢量插画"
### 项目结构
```
pixabay_mcp/
├── src/
│ └── pixabay_mcp/
│ ├── __init__.py
│ └── server.py # MCP 服务器主实现
├── pyproject.toml # 项目配置文件
├── uv.lock # 依赖锁定文件
├── README.md # 本文件
├── LICENSE # MIT 许可证
└── .gitignore # Git 忽略规则
```
### uv 和 uvx 的区别
**uv** - 通用 Python 项目管理工具
- 用于开发、安装包、运行脚本
- 需要手动管理虚拟环境
- 适合本地开发和调试
**uvx** - 快速执行工具
- 直接从 PyPI 或 Git 仓库运行包
- 自动管理隔离环境
- 无需手动安装,开箱即用
- 适合快速部署和分享
### 许可证
MIT 许可证
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: one searches for images and the other searches for videos. There is no overlap in functionality, and the descriptions explicitly differentiate between media types, making it impossible for an agent to confuse them.
Both tools follow a consistent verb_noun pattern with 'search_' prefix followed by the media type (images/videos). The naming is perfectly uniform and predictable across the tool set.
With only two tools, this server feels thin for a media search service. While search is core, there are obvious gaps like fetching specific media by ID, getting trending content, or managing downloads. The count is too low for comprehensive coverage of the Pixabay domain.
The tool surface is severely incomplete for a Pixabay API server. It only provides search functionality, missing essential operations like retrieving specific images/videos by ID, getting user uploads, or accessing categories/trends. Agents will hit dead ends when trying to perform basic media retrieval tasks beyond searching.