Cursor → Claude Code: Migrate Your AI Memory (2026 Method)

Cursor's .cursorrules, .cursor/rules/*.mdc, and Notepads don't auto-import into Claude Code. You'll consolidate your rules into a CLAUDE.md at the repo root (and ~/.claude/CLAUDE.md for global), translate Notepads into custom slash commands, and rewire MCP servers in Claude Code's settings. Plan 15–30 minutes per repo. A shared MCP-based memory layer like MemoryLake lets both tools read the same source.

Three drivers in 2026:

• Terminal-native workflow. Engineers shipping multi-repo refactors often prefer Claude Code's CLI flow over an IDE.
• Long-running agent tasks. Claude Code handles longer multi-step tasks with deeper context per session.
• MCP-first defaults. Claude Code's first-class MCP support fits teams already standardizing on MCP servers for tools and memory.

Different surface, similar intent.

Cursor memory spans `.cursorrules` (legacy single-file project rules), *`.cursor/rules/.mdc` (structured per-rule format with metadata), Notepads (saved reusable prompts), and user-level Rules for AI** in Settings.

Claude Code memory centers on `CLAUDE.md` at the repo root (project-level guidance), `~/.claude/CLAUDE.md` (user-global guidance), custom slash commands (reusable prompts saved as .md files), and MCP servers configured in Claude Code settings.

A .cursorrules file usually becomes a CLAUDE.md. .mdc rules collapse into sections of CLAUDE.md. Notepads become custom slash commands.

There's no Cursor export command; rules are files you copy.

1. Locate your repo's rules. Check .cursorrules and .cursor/rules/.
2. Read each `.mdc` rule's frontmatter and body. Note which apply always and which apply on certain globs.
3. Export your Notepads. Copy each Notepad's name and content into a notepads.md.
4. Capture user-level Rules for AI. Settings → Rules for AI. Paste into cursor-user-rules.txt.

End state: a cursor-export/ folder per repo containing the rules files (or contents), notepads.md, and cursor-user-rules.txt.

Claude Code expects a CLAUDE.md and a .claude/commands/ directory.

1. Create `CLAUDE.md` at the repo root. Start with your .cursorrules content. Add a clear section heading for each .mdc rule cluster.
2. Add glob notes inline. Where a .mdc had a globs pattern, mention it in prose ("for files under src/api/...") so the model applies the guidance contextually.
3. Add user-global guidance to `~/.claude/CLAUDE.md`. Paste your Cursor user-level Rules for AI.
4. Translate Notepads into slash commands. For each Notepad, create .claude/commands/<name>.md containing the prompt body. Claude Code exposes it as /<name>.
5. Configure MCP servers. Edit Claude Code settings to add any MCP servers you previously used in Cursor. Reuse endpoint URLs and Bearer tokens.
6. Probe. Run a small task in Claude Code that depends on a moved rule.

• `.mdc` frontmatter semantics. Claude Code reads CLAUDE.md as prose; glob-based selective application doesn't transfer automatically.
• IDE-specific affordances. Inline Tab completion and Cursor-style chat are IDE features, not CLI ones.
• Per-IDE chat history. Cursor's chats stay in the IDE.
• Ongoing sync. New .cursorrules edits next week won't appear in CLAUDE.md unless you redo the copy.

If you work across IDEs and the terminal, per-tool rules files drift fast. Worse, anything broader than a single repo (cross-repo conventions, team standards) lives in two places.

MemoryLake holds those rules and standards once and exposes them through MCP. Both Cursor and Claude Code support MCP, so the same project context flows into both from a single endpoint.

• One source of truth. Update a rule once; both tools see the change.
• Cross-repo standards. Team conventions live above any single repo.
• Drop-in for the next tool. Add Windsurf or a future agent with a config change.

Step 1: Create a project and load your context

Sign in to MemoryLake, open Project Management, and click Create Project. Name it "Cursor ↔ Claude Code shared rules." Drag reference docs (style guides, ADRs as PDF, Word, Markdown, or images) into the Document Drive under My Space, then open the Documents Tab and click Add Documents. Paste your .cursorrules, .mdc bodies, Notepads, and user-level rules into the Memories Tab via Add Memory.

Step 2: Generate an MCP Server endpoint

Open the MCP Servers Tab inside the project, click Add MCP Server, describe it (e.g., "Shared dev rules"), and click Generate. MemoryLake returns a Key ID, a Secret, and an Endpoint URL. Copy the Secret immediately — it is shown only once.

Step 3: Point both tools at the endpoint

In Cursor's MCP configuration (e.g., .cursor/mcp.json), add a MemoryLake server entry with the endpoint URL and the Secret as a Bearer token. In Claude Code's settings, add the same MCP server. Restart Cursor and re-open Claude Code.