Summary
Relay launched as a shared project-memory product that captures context from browser-based AI chats and makes the same brief available to coding agents through MCP. It is positioning itself as a continuity layer across ChatGPT, Claude, Gemini, Codex, Cursor, Windsurf, and other agent surfaces.
What changed
Relay publicly launched a browser-plus-MCP workflow that captures project decisions, tasks, and constraints from AI chats and syncs them into a shared project brief for IDE and CLI agents.
Why it matters
Context fragmentation is becoming a major workflow tax as teams spread work across multiple AI tools. Relay is betting that memory should live above any single model or interface, with MCP acting as the bridge between consumer chat apps and coding-agent environments. That makes shared project memory a product category, not just an ad hoc pattern.
Evidence excerpt
Relay says it captures what matters from AI chats, keeps a living project brief, and exposes that context to coding agents through MCP across tools like Claude Code, Cursor, Windsurf, and Codex.