Export Manus Memory and Import to Claude — 2026 Guide

Manus has no official export to Claude. You'll copy each task's prompts, attached knowledge, and notable outputs by hand into a Claude Project — System Prompt for the task pattern, Project Knowledge for the supporting documents and prior-run summaries. Plan 25–45 minutes per task pattern. A shared MCP-based memory layer like MemoryLake lets both tools read the same source.

Three drivers in 2026:

• Conversational depth. Claude's chat surface fits collaborative analysis better than an agent runner for many tasks.
• MCP-native client. Claude Desktop's MCP support enables sharing tools and memory cleanly.
• Cost predictability. Long agent runs can be expensive; Claude pricing is more predictable per-conversation.

Different surfaces.

Manus memory lives across Tasks (defined task patterns with prompts), Knowledge (uploaded reference material per task), and the run history of past task executions.

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

A Manus task usually becomes a Claude Project. Task prompts become System Prompts. Task Knowledge becomes Project Knowledge. Important prior-run outputs become Project Knowledge text excerpts.

Manus exposes tasks and knowledge through the UI.

1. List your active task patterns. For each task, copy its name, description, and prompt template into a text file.
2. Download attached Knowledge files. Save originals where Manus exposes them.
3. Capture notable run outputs. For runs that produced reusable artifacts (research summaries, plans, reports), save the output as markdown.
4. Note connected accounts. List any external accounts the task relied on; you'll need parallel access from Claude (often via MCP servers).

End state: one folder per Manus task with prompt.md, original Knowledge files, run output excerpts, and connections.md.

Claude expects per-Project Knowledge.

1. Create a Project per Manus task. Projects → Create Project. Name it after the task.
2. Paste the task prompt as the System Prompt. Rewrite phrasing for an interactive chat rather than an agent run ("I'll help you..." instead of "Execute this task...").
3. Upload Knowledge files. Click Add Content → Upload Files.
4. Add prior-run output excerpts as Project Knowledge text. Paste salient summaries so Claude can reference your past work.
5. Rebuild agentic capability via MCP servers (if needed). For tools Manus invoked, add equivalent MCP servers to Claude Desktop's MCP config.
6. Probe. Open the Project and run the equivalent task interactively.

Claude does not import Manus run history directly; only what you bring across survives.

• Fully autonomous task execution. Claude is interactive by default; multi-step autonomous loops feel different.
• Manus-side scheduling. Recurring task scheduling needs a parallel mechanism (cron, GitHub Actions, etc.).
• Connected account convenience. Each external account needs reconnecting through MCP servers on Claude's side.
• Ongoing sync. New Manus runs next week won't update Claude unless you re-import excerpts.

If you keep Manus for scheduled autonomous work and Claude for interactive analysis on the same outputs, per-tool drift starts immediately.

MemoryLake holds canonical task knowledge and rules once and exposes them through MCP. Claude Desktop reads MCP natively; Manus tasks can pull from MemoryLake through its REST endpoint.

• One source of truth. Update once; both tools see the change.
• Standard file formats. PDFs, Word, Excel, PowerPoint, Markdown, and images live in MemoryLake's Document Drive as-is.
• Drop-in for the next AI. Add ChatGPT or a different agent runner 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 after the Manus task. Drag Knowledge 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 task prompts and important run-output summaries 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., "Manus + Claude 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 tools 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. For Manus, configure your task to call the REST endpoint with the Bearer token to pull or store task knowledge at runtime.