I use ChatGPT Projects as long-running collaborative workspaces rather than isolated conversations. When returning to a complex project after several days, I have repeatedly encountered situations where previously completed work or rejected approaches are effectively rediscovered.
This led me to the following feature proposal:
**Feature request: “Project Context Synchronization on Open” (ChatGPT Projects)**
When a user opens a Project, ChatGPT should automatically (optionally + asynchronously) scan the Project’s existing chats and sources (files/links/instructions) and generate/refresh a compact internal “Project State” so the user doesn’t have to maintain it manually. The state should include: Objective, key definitions, decisions made, completed work, rejected approaches (with rationale), open items/next steps, and the current/last working state. This would reduce repetitive re-briefing after time away and make Projects function more like persistent collaborative workspaces rather than separate threads.
It would be great if this would be fixed in some ways.