AI Skill Hub 强烈推荐：多智能体AI系统构建工作坊 是一款优质的AI工具。AI 综合评分 8.2 分，在同类工具中表现稳健。如果你正在寻找可靠的AI工具解决方案，这是一个值得深入了解的选择。
基于MCP的开源工作坊，教授从零开始构建多智能体AI系统。提供Deep Research Agent等实战案例，适合想深入学习AI Agent架构和工作流设计的开发者和AI从业者。
多智能体AI系统构建工作坊 是一款基于 Python 开发的开源工具，专注于 多智能体系统、MCP工具、深度研究代理 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
基于MCP的开源工作坊，教授从零开始构建多智能体AI系统。提供Deep Research Agent等实战案例，适合想深入学习AI Agent架构和工作流设计的开发者和AI从业者。
多智能体AI系统构建工作坊 是一款基于 Python 开发的开源工具，专注于 多智能体系统、MCP工具、深度研究代理 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
# 方式一：pip 安装（推荐）
pip install designing-real-world-ai-agents-workshop
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install designing-real-world-ai-agents-workshop
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/iusztinpaul/designing-real-world-ai-agents-workshop
cd designing-real-world-ai-agents-workshop
pip install -e .
# 验证安装
python -c "import designing_real_world_ai_agents_workshop; print('安装成功')"
# 命令行使用
designing-real-world-ai-agents-workshop --help
# 基本用法
designing-real-world-ai-agents-workshop input_file -o output_file
# Python 代码中调用
import designing_real_world_ai_agents_workshop
# 示例
result = designing_real_world_ai_agents_workshop.process("input")
print(result)
# designing-real-world-ai-agents-workshop 配置文件示例（config.yml） app: name: "designing-real-world-ai-agents-workshop" debug: false log_level: "INFO" # 运行时指定配置文件 designing-real-world-ai-agents-workshop --config config.yml # 或通过环境变量配置 export DESIGNING_REAL_WORLD_AI_AGENTS_WORKSHOP_API_KEY="your-key" export DESIGNING_REAL_WORLD_AI_AGENTS_WORKSHOP_OUTPUT_DIR="./output"
| Requirement | Check | Install |
|---|---|---|
| Python 3.12+ | python --version | uv python install 3.12 or [python.org](https://www.python.org/downloads/) |
| uv 0.7+ | uv --version | curl -LsSf https://astral.sh/uv/install.sh \| sh ([docs](https://docs.astral.sh/uv/getting-started/installation/)) |
| GNU Make | make --version | Pre-installed on macOS/Linux. Windows: choco install make |
| Google API Key | — | [aistudio.google.com/apikey](https://aistudio.google.com/apikey) (required — all LLM calls use Gemini) |
| Opik account | — | [comet.com/site/products/opik](https://www.comet.com/site/products/opik/) (optional, for observability and evals) |
The workshop includes an LLM-as-judge evaluation pipeline. Instead of manually reviewing each generated post, an LLM scores them against quality criteria (structure, tone, accuracy). Opik tracks these scores across runs so you can measure whether prompt or pipeline changes actually improve output quality.
make eval-dev # LLM judge on dev split
make eval-test # LLM judge on test split
make eval-online # Generate + judge posts on the flyEach command automatically uploads the dataset to Opik before running. To upload without evaluating (e.g., to browse in the Opik UI), use make upload-eval-dataset.
A hands-on workshop, presented at AI Engineering Conference Europe, building a multi-agent AI system with two MCP servers: a Deep Research Agent and a LinkedIn Writing Workflow. Both connected to a harness like Claude Code or Cursor.
🎬 Full workshop available on YouTube ↓
<a href="https://www.youtube.com/watch?v=mYSRn6PC1mc"> <img src="https://img.youtube.com/vi/mYSRn6PC1mc/maxresdefault.jpg" alt="Watch the video" style="width:100%; max-width:600px;"> </a>
📑 Slides here.
----
Deep Research Agent — An MCP server that runs deep research using Gemini with Google Search grounding and native YouTube video analysis:
user topic → [deep_research] × N → analyze_youtube_video (if URLs) → [deep_research gap-fill] → compile_research → research.mdLinkedIn Writing Workflow — An MCP server that generates LinkedIn posts with an evaluator-optimizer loop:
research.md + guideline → generate post → [review → edit] × N → post.md → generate imageBoth servers expose tools, resources, and prompts via the Model Context Protocol, letting any MCP-compatible harness orchestrate the workflow.
<img src="media/architecture.png" alt="End-to-end workflow architecture" width="800"/>
Patterns and concepts you'll learn:
<img width="1400" height="1380" alt="system_architecture" src="https://github.com/user-attachments/assets/5507d5dd-5809-4e01-bcf3-a6de980bc773" />
Assumes working Python knowledge and basic familiarity with LLMs.
git clone https://github.com/iusztinpaul/designing-real-world-ai-agents-workshop.git
cd designing-real-world-ai-agents-workshop
cp .env.example .env # add your GOOGLE_API_KEY (+ optional OPIK_API_KEY)
uv sync
Note: If you don't have Python 3.12+, uv can install it for you:uv python install 3.12, then re-runuv sync.
make test-end-to-end # runs research + writing pipeline end-to-end
If it completes without errors, you're good to go.
Three ways to use this repo. Pick the mode that fits the time you have. Or work through all three in order, since each builds on the last:
implement_yourself/, a stripped-down skeleton prepared with 25 pre-groomed tickets and a custom /implement Claude Code skill that orchestrates SWE and Tester agents in a loop, ticket by ticket, until the directory matches src/. See implement_yourself/README.md for the kickoff guide.No cheating, by design.implement_yourself/is a self-contained project. Open your harness (Claude Code, Cursor, …) directly in that folder (not at the repo root) so its working directory is scoped to the skeleton. The agents can't see the reference implementation in../src/, can't grep it, can't read its files. You get a real build, not a copy-paste.
Here's a real run through the full pipeline — from a topic seed to a published-ready LinkedIn post with an AI-generated image.
- Stop Overengineering: Workflows vs AI Agents Explained (YouTube) - From 12 Agents to 1 (DecodingAI article)
#### 2. Deep Research Agent produces `research.md`
The agent runs multiple Gemini-grounded search queries and analyzes YouTube videos, then compiles everything into a structured research brief with sources.
> The full research.md for this example is ~20k tokens across 2 queries and 1 video transcript.
#### 3. Write a guideline
A short brief describing the post angle, audience, and key points:
高质量教学工程项目，结合MCP标准和多智能体范式，提供完整实战框架。Star数适中但质量突出，适合作为Agent开发学习资源。
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一，可自由商用、修改、分发，仅需保留版权声明。
总体来看，多智能体AI系统构建工作坊 是一款质量优秀的AI工具，在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态，建议收藏备用，结合自身场景选择合适时机引入使用。
| 原始名称 | designing-real-world-ai-agents-workshop |
| Topics | 多智能体系统MCP工具深度研究代理AI工作流Python实战 |
| GitHub | https://github.com/iusztinpaul/designing-real-world-ai-agents-workshop |
| License | MIT |
| 语言 | Python |
收录时间：2026-05-25 · 更新时间：2026-05-25 · License：MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。