Universal persistent memory across all Claude surfaces

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

Lack of Universal Memory per User/Team

Proposed Solution

Hi Anthropic team,

I'd like to suggest a feature that I think would significantly improve the experience for power users: universal, persistent memory that works across all Claude interfaces — claude.ai (web), Claude desktop (Cowork), Claude Code (CLI/VPS), and the VS Code extension.
The problem today:
Each interface has its own isolated context. A user who works across multiple environments has to re-explain their preferences, projects, and background in every single session. This creates friction and limits how useful Claude can be as a long-term assistant.
The proposal:
A unified memory layer — a structured file (e.g. JSON or Markdown) containing user preferences, ongoing projects, and key facts — that every Claude interface can read from and write to.
The key design idea — user-defined storage:
Rather than forcing a single centralized store, let users choose where their memory lives:
Anthropic cloud (zero-config default)
Their own S3/R2/Blob storage (for enterprise/compliance needs)
A self-hosted server or VPS
A private Git repository (with built-in versioning and history)
This approach respects privacy, satisfies compliance requirements, and gives technical users full control.
Why this matters:
For developers and professionals working across web, desktop, and CLI environments simultaneously, this would transform Claude from a per-session tool into a genuine long-term collaborator.

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

CLI commands and flags

Use Case Example

_No response_

Additional Context

_No response_