How to Build a QwenPaw Agent Workspace with Custom Skills, Model Providers, Console Access, and Streaming API Testing
English summary
This tutorial demonstrates a complete QwenPaw agent workspace setup in Google Colab. It covers installation, environment configuration, and authentication with a randomly generated password. The setup automatically configures an LLM provider from Colab secrets (OpenAI, OpenRouter, DashScope, DeepSeek, or Gemini) and creates a custom research_brief skill along with demo knowledge files. The QwenPaw Console is launched on a configurable port, made accessible via Colab proxy and optionally a Cloudflare tunnel, and a streaming chat API client is used to test the agent's responses. The result is a reproducible, secure, and extensible local-first agent platform for research and automation workflows.
Chinese summary
本教程展示了在 Google Colab 中搭建完整的 QwenPaw 智能体工作区。流程包括安装、环境配置和随机密码认证。系统自动从 Colab 密文配置大语言模型提供商（OpenAI、OpenRouter、DashScope、DeepSeek 或 Gemini），并创建自定义 research_brief 技能以及演示知识文件。QwenPaw 控制台在可配置端口启动，通过 Colab 代理和可选的 Cloudflare 隧道暴露访问，随后使用流式聊天 API 客户端测试智能体的响应。最终得到一个可复现、安全且可扩展的本地优先智能体平台，适用于研究与自动化流程。
Key points
A QwenPaw workspace is initialized in Google Colab with working directories, authentication, and helper scripts for process management.
在 Google Colab 中初始化 QwenPaw 工作区，配置工作目录、认证和进程管理辅助脚本。
Model providers are auto-detected from Colab secrets or environment variables, supporting OpenAI, OpenRouter, DashScope, DeepSeek, and Gemini with provider-specific configuration files.
从 Colab 密文或环境变量自动检测模型提供商，支持 OpenAI、OpenRouter、DashScope、DeepSeek 和 Gemini，并生成各自的配置文件。
A custom research_brief skill is defined with a markdown instruction file, and demo knowledge files are placed in the workspace to give the agent local context.
通过 Markdown 指令文件定义自定义 research_brief 技能，并在工作区放置演示知识文件，为智能体提供本地上下文。
The QwenPaw Console server is launched on a configurable port, with credentials printed, and optionally exposed through a Cloudflare tunnel for public access.
在可配置端口启动 QwenPaw 控制台服务器，打印登录凭据，并可选通过 Cloudflare 隧道暴露以供公开访问。
A streaming chat API client is provided to test the agent, sending a prompt and processing server-sent events to display the generated response.
提供流式聊天 API 客户端测试智能体，发送提示并处理服务器发送事件以显示生成的回复。