What problem does it solve?
Designing AI agent workflows by hand-writing JSON or Markdown configuration is error-prone and slow. This Skill lets you create and modify visual workflows in CC Workflow Studio through natural conversation, using the built-in MCP server to read the schema, inspect the current canvas, and apply changes directly.
Core Features & Use Cases
- Conversational workflow creation: Describe the workflow you want and the Skill generates valid workflow JSON using built-in sub-agents (explore, plan, general-purpose).
- Incremental canvas editing: Apply structural changes with
apply_workflow or make token-efficient partial updates to existing nodes with update_nodes.
- Visual organization with group nodes: Group related nodes into labeled containers with defined dimensions and relative child positioning, without affecting execution.
- Use Case: Ask the assistant to build a workflow that fetches data, plans a transformation, and summarizes results; it fetches the schema, generates the node graph, applies it to the canvas, and iterates based on your feedback.
Quick Start
Ask the assistant to create a new workflow in CC Workflow Studio with a fetch-data sub-agent and a summarize sub-agent connected in sequence.