Krea Nodes workflows can now be co-edited and run live by a team, with per-user undo and shared run status, replacing link-and-copy handoffs of generative pipelines.
Key takeaways · AI-distilled
Krea Nodes now lets several people edit and run the same workflow graph at once, with live cursors and selections, shared runs whose outputs land on every screen, and an undo stack that reverts only your own changes.
Concurrent edits follow a simple rule: changes to different nodes or fields all apply, but when two people edit the same field, such as one prompt, the last edit wins.
Whoever starts a run pays for it in their active workspace. A chain keeps running from the submitter's tab even after they switch tabs, while other collaborators only watch the results arrive.
View sharing works on every plan, but co-editing and live presence require Krea Enterprise. If a workspace leaves Enterprise, its edit grants act as view grants until it returns.
Viewers see the graph and its inputs but not outputs or run history, and can duplicate a project, so a public link works as a reusable template. Krea's AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → and MCPThe Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.Full definition → clients edit through the same operations as people.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
