designing-real-world-ai-agents-workshopMCP Server37/100 via “end-to-end workflow orchestration from research to published content”
Hands-on workshop: Build a multi-agent AI system from scratch — Deep Research Agent + Writing Workflow served as MCP servers. Includes code, slides, and video
Unique: Exposes the entire research-to-content pipeline as a single Claude Code skill, enabling non-technical users to run complex multi-agent workflows without understanding MCP or agent architecture. Filesystem-based state persistence allows inspection and manual intervention at any stage.
vs others: More complete than individual agent tools because it handles the full pipeline (research + writing + evaluation + images), and more accessible than custom orchestration code because it's exposed as a Claude Code skill with natural language invocation.