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

The Claude chat product and Claude Code don't share memory automatically. You'll download each Project's Knowledge files, copy each System Prompt into a CLAUDE.md at the matching repo root (or ~/.claude/CLAUDE.md for global guidance), and convert reusable prompts into slash commands under .claude/commands/. Plan 15–25 minutes per repo. A shared MCP-based memory layer like MemoryLake lets both clients read the same source.

Three drivers in 2026:

• Repo-native execution. Claude Code reads, edits, and runs commands in your repo without copy-paste.
• Multi-step coherence. Long refactors and migrations stay coherent across many turns.
• MCP and tool integration. First-class MCP and shell access fit teams shipping multi-file work.

Same model, different surfaces.

Claude chat memory lives inside Projects. Each Project has its own Project Knowledge (uploaded files and pasted text) and an optional System Prompt.

Claude Code memory centers on `CLAUDE.md` at the repo root, `~/.claude/CLAUDE.md` (user-global), custom slash commands under .claude/commands/, and MCP servers configured in Claude Code settings.

A Project's System Prompt becomes a CLAUDE.md. Project Knowledge becomes repo docs/ references. Reusable prompts become slash commands.

Claude has no Project-export bundle.

1. Capture each Project's System Prompt. Open the Project → Project Instructions. Copy the contents into a text file labelled with the Project name.
2. Download Project Knowledge files. Click each file and download originals. Re-upload from local copies if available.
3. Copy pasted-text knowledge. Save into a notes.md per Project.
4. List recurring prompts. Capture frequent question patterns in a prompts/ directory, one file per pattern.

End state: one folder per Claude Project with the System Prompt, original files, notes.md, and prompts/.

Claude Code expects plain markdown.

1. Map each Project to a repo. If a Project corresponds to a single codebase, target that repo. For broader Projects, pick a primary repo plus user-global guidance.
2. Create `CLAUDE.md` at the repo root. Paste the Project System Prompt, restructured for repo-specific concerns.
3. Add Knowledge as repo docs. Place downloaded files under docs/ and reference them inside CLAUDE.md.
4. Write `~/.claude/CLAUDE.md`. Paste cross-Project guidance you want available everywhere.
5. Create `.claude/commands/<name>.md` for each reusable prompt. Claude Code exposes them as /<name>.
6. Re-add MCP servers. Open Claude Code settings and add the MCP servers you used in Claude Desktop with their endpoint URLs and Bearer tokens.
7. Probe. Run a small slash command or task in Claude Code.

• Chat thread continuity. Claude chat conversations stay in the Claude UI.
• Per-Project boundary. A CLAUDE.md reads alongside the repo content; the explicit Project wall becomes less visual.
• Image and file-pull conversational habits. Terminal flow differs from chat flow.
• Ongoing sync. New Project Knowledge in Claude next week won't appear in CLAUDE.md unless you redo the copy.

If you use Claude chat for design and Claude Code for shipping, files drift the moment you finish migrating. Cross-repo standards end up in two places.

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

• One source of truth. Update once; both surfaces see the change.
• Cross-repo standards. Team conventions live above any single repo.
• Drop-in for the next tool. Add Cursor or Windsurf 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 "Claude ↔ Claude Code shared context." Drag your downloaded files (PDF, Word, Excel, PowerPoint, Markdown, or images) into the Document Drive under My Space, then open the Documents Tab and click Add Documents. Paste each System Prompt and your reusable prompts 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., "Claude + Claude Code bridge"), 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 clients at the endpoint

Add MemoryLake to Claude Desktop's MCP config with the endpoint URL and the Secret as a Bearer token, then restart Claude. In Claude Code's settings, add the same MCP server. Re-open Claude Code.