bfc2743f3a
- Disable terminal server tools from being exposed to AI - Add RichUI theme-aware CSS variables for better text contrast - Update version to 0.12.1 across all documentation files
124 lines
3.5 KiB
Markdown
124 lines
3.5 KiB
Markdown
# Pipe 插件
|
||
|
||
Pipe 插件用于创建自定义模型集成或转换 LLM 响应，会在 OpenWebUI 的模型下拉中显示。
|
||
|
||
## 什么是 Pipes？
|
||
|
||
Pipes 可以用于：
|
||
|
||
- :material-api: 连接外部 AI API（Gemini、Claude 等）
|
||
- :material-robot: 创建自定义模型封装
|
||
- :material-cog-transfer: 变换请求与响应
|
||
- :material-middleware: 实现中间件逻辑
|
||
|
||
---
|
||
|
||
## 可用的 Pipe 插件
|
||
|
||
- [GitHub Copilot SDK](github-copilot-sdk.zh.md) (v0.12.1) - GitHub Copilot SDK 官方集成。具备**工作区安全隔离**、**零配置工具桥接**与**BYOK (自带 Key) 支持**。**v0.12.1 更新：禁用终端工具 AI 调用、RichUI 主题感知 CSS 变量**。[查看深度架构解析](github-copilot-sdk-deep-dive.zh.md) | [**查看进阶 实战教程**](github-copilot-sdk-tutorial.zh.md) | [**查看详细使用手册**](github-copilot-sdk-usage-guide.zh.md)。
|
||
- **[实战案例：GitHub 100 Star 增长预测](star-prediction-example.zh.md)** - 展示如何使用 GitHub Copilot SDK Pipe 结合 Minimax 2.1 模型，自动编写脚本分析 CSV 数据并生成详细的项目增长报告。
|
||
- **[实战案例：视频高质量 GIF 转换与加速](video-processing-example.zh.md)** - 演示模型如何通过底层 FFmpeg 工具对录屏进行加速、缩放及双阶段色彩优化处理。
|
||
|
||
---
|
||
|
||
## Pipe 工作原理
|
||
|
||
```mermaid
|
||
graph LR
|
||
A[User selects Pipe as Model] --> B[Pipe receives request]
|
||
B --> C[Transform/Route request]
|
||
C --> D[External API / Custom Logic]
|
||
D --> E[Return response]
|
||
E --> F[Display to User]
|
||
```
|
||
|
||
### `pipes` 方法
|
||
|
||
定义此 Pipe 提供的模型列表：
|
||
|
||
```python
|
||
def pipes(self):
|
||
return [
|
||
{"id": "my-model", "name": "My Custom Model"},
|
||
{"id": "my-model-fast", "name": "My Custom Model (Fast)"}
|
||
]
|
||
```
|
||
|
||
### `pipe` 方法
|
||
|
||
负责处理实际请求：
|
||
|
||
```python
|
||
def pipe(self, body: dict) -> Generator:
|
||
# 处理请求
|
||
messages = body.get("messages", [])
|
||
|
||
# 调用外部 API 或自定义逻辑
|
||
response = call_external_api(messages)
|
||
|
||
# 返回响应（可流式）
|
||
return response
|
||
```
|
||
|
||
---
|
||
|
||
## 快速安装
|
||
|
||
1. 下载需要的 Pipe `.py` 文件
|
||
2. 前往 **Admin Panel** → **Settings** → **Functions**
|
||
3. 上传并配置 API Key
|
||
4. 该 Pipe 会作为模型选项出现
|
||
|
||
---
|
||
|
||
## 开发模板
|
||
|
||
```python
|
||
"""
|
||
title: My Custom Pipe
|
||
author: Your Name
|
||
version: 1.0.0
|
||
description: Description of your pipe plugin
|
||
"""
|
||
|
||
from pydantic import BaseModel, Field
|
||
from typing import Generator, Iterator, Union
|
||
|
||
class Pipe:
|
||
class Valves(BaseModel):
|
||
API_KEY: str = Field(
|
||
default="",
|
||
description="API key for the external service"
|
||
)
|
||
API_URL: str = Field(
|
||
default="https://api.example.com",
|
||
description="API endpoint URL"
|
||
)
|
||
|
||
def __init__(self):
|
||
self.valves = self.Valves()
|
||
|
||
def pipes(self) -> list[dict]:
|
||
"""Define available models."""
|
||
return [
|
||
{"id": "my-model", "name": "My Custom Model"},
|
||
]
|
||
|
||
def pipe(
|
||
self,
|
||
body: dict
|
||
) -> Union[str, Generator, Iterator]:
|
||
"""Process the request and return response."""
|
||
messages = body.get("messages", [])
|
||
model = body.get("model", "")
|
||
|
||
# 自定义逻辑
|
||
# 返回值可以是：
|
||
# - str：单次响应
|
||
# - Generator/Iterator：流式响应
|
||
|
||
return "Response from custom pipe"
|
||
```
|
||
|
||
更多细节见 [插件开发指南](../../development/plugin-guide.md)。
|