[FEATURE] Portable project memory across machines (multi-device workflow)

[hc-7000]

Preflight Checklist

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

Problem Statement

When working on the same project from multiple computers (e.g. a workstation and a laptop), Claude Code's project memory doesn't carry over. The ~/.claude/projects/ directory stores memories under paths derived from the absolute project path (e.g. -Users-xxx-Sync-Project-yyy), which means:

1. Different usernames between machines = different memory subdirectory
2. Different mount points for the same synced folder = different memory subdirectory
3. Settings, plans, and session data are all local to one machine

For users who sync their project folders (via iCloud, Syncthing, Dropbox, etc.) and switch between machines daily, this means Claude Code loses all accumulated project context each time.

Proposed Solution

A few possible approaches:

• Project-relative memory: Allow storing .claude/memory/ inside the project directory itself (similar to how .claude/settings.json can be project-local), so it syncs naturally with the project.
• Project identity by name, not path: Resolve memory by a project identifier (e.g. git remote URL, or a .claude/project-id file) rather than the absolute filesystem path.
• Memory sync/export: A /memory export and /memory import command to make portability explicit.

Environment

• macOS, two machines (workstation + laptop)
• Projects synced via cloud sync
• Same user, same projects, same (or different) absolute paths

Alternative Solutions

Manually copying MEMORY.md into the project folder alongside CLAUDE.md. This works but loses the automatic memory management — you have to remember to keep it updated.

Priority

Medium - Would be very helpful

Feature Category

Configuration and settings

Use Case Example

No response

Additional Context

No response

[github-actions]

Found 3 possible duplicate issues:

1. [FEATURE] Custom Project Identifier for Memory Directory #23321
2. Allow user-defined named projects that map to multiple directories #24980
3. Auto memory should be shared across git worktrees of the same repository #24382

This issue will be automatically closed as a duplicate in 3 days.

• If your issue is a duplicate, please close it and 👍 the existing issue instead
• To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

[StealUrKill]

Have you seen this project?

https://github.com/kneetworks/claude-sync

[Daniel-de-Oliveira-Trindade]

This matches a core problem I ran into too — different absolute paths between machines breaking session continuity. I built claude-context-sync which solves the path issue by converting absolute paths to template variables (${PROJECTS}, ${HOME}, etc.) during export and restoring them on import.

It doesn't sync memory in-place inside the project folder, but it does let you carry full session context (messages, file history, todos) between machines via a Git repo, even when paths differ. Might be useful as a workaround until a native solution ships!

[m13v]

We solved this for our own workflow with user-memories - a standalone knowledge extraction system that syncs across machines.

The approach: instead of syncing Claude Code's internal project memory files, we extract knowledge into a SQLite database with semantic embeddings. The DB file is small (usually under 50MB even for large projects) and easy to sync via git, Dropbox, or rsync. On each machine, a CLAUDE.md instruction points Claude to query this shared knowledge base.

The key advantage over syncing ~/.claude/projects/ directly: semantic search means the right context surfaces regardless of which machine created it. A memory created on your workstation about "the auth module uses JWT with RS256" will match when you are on your laptop asking about authentication, even if the path-derived project key differs.

For the path-derivation problem specifically, we hash the git remote URL instead of the local filesystem path. Two clones of the same repo on different machines produce the same project key. This would be a relatively simple fix in Claude Code's own memory system.

[m13v]

The user-memories project that handles cross-machine knowledge sync: https://github.com/m13v/user-memories/blob/main/user_memories/db.py

Embedding-based semantic retrieval so memories match by meaning, not path: https://github.com/m13v/user-memories/blob/main/user_memories/embeddings.py