One memory for n8n, and every tool you will use
Automations store and recall from the same Hypermemory graph as Claude and your desktop app, so ops and ad-hoc chats never disagree about what happened.
Why This Works
Hypermemory is accessed via MCP and the CLI, the same memory graph either way. n8n workflows connect via the MCP node; you can also call the CLI from script steps. hm_recall pulls what ChatGPT or Claude already stored yesterday.
Quick Setup
MCP URL: https://api.hypermemory.io/mcp
- Create a Hypermemory account and copy your API key and graph ID from the app.
- Add an MCP Client node in n8n and paste the Hypermemory MCP URL.
- Authenticate with your API key (or OAuth where supported).
- Call hm_store and hm_recall in your workflow, the same graph as Claude and your other tools.
Available Tools
| Tool | Description |
|---|---|
| hm_find_related | Traverse connected knowledge from a starting node |
| hm_forget | Delete a node with optional cascade |
| hm_get_overview | Graph stats, top nodes, and type distribution |
| hm_ingest | Decompose dense text into graph entities in one pass |
| hm_list_files | Query uploaded files by type or name |
| hm_list_orphans | List disconnected nodes for cleanup |
| hm_recall | Hybrid search across your graph (BM25, vector, edges) |
| hm_store | Create a memory node; background enrichment runs automatically |
| hm_timeline | Temporal recall by period or date range |
| hm_timeline_write | Write an explicit diary-style timeline entry |
| hm_update | Update a node; re-embeds when the description changes |
| hm_upload_file | Upload a file into memory (summary + graph node) |
Use Case
Ticket triage stores the fix pattern with hm_store. When a similar ticket arrives, the workflow hm_recalls the prior thread, and your PM sees the same story in Claude without opening n8n. One memory, multiple surfaces.