MemorySync
Durable, inspectable memory for AI agents and coding assistants.
What it can do
- Search Memory: Search the user's stored memory for context relevant to a question and return the best matches. Use this before answering anything that depends on what the user told you previously.
- Add Memory: Store a durable fact or preference about the user. Text is run through extraction, so conversational filler is discarded and near-duplicates are merged; a skipped result is normal and not
- List Memories: List the user's most recent memories without searching.
What data it sees
Do you need an account
No: the server works without sign-in
Durable, inspectable memory for AI agents and coding assistants. Scopes decisions and context across sessions.
How to connect
How to connect MemorySync to Claude
The server address to paste into Claude settings:
https://memorysync--memorysync-io.run.tools
- Open Claude (claude.ai or the desktop app).
- Go to Settings → Connectors.
- Click Add custom connector.
- Paste the server address copied below into Remote MCP server URL and click Add.
- In a chat, click + → Connectors and switch the new connector on.
Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.
Authorization
After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.
How to connect MemorySync to ChatGPT
The server address to paste into ChatGPT settings:
https://memorysync--memorysync-io.run.tools
- Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
- Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
- Open Settings → Connectors and click Create.
- Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
- Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
- In a new chat open + → Apps/Connectors and enable the connector.
OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.
Authorization
On first use ChatGPT opens the service's sign-in window. Sign in and approve access.
How to connect MemorySync to Cursor
The server address to paste into Cursor settings:
https://memorysync--memorysync-io.run.tools
Fastest: click Open in Cursor below and confirm the prompt.
Manually:
- In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
- Paste the JSON copied below into
~/.cursor/mcp.json(per project:.cursor/mcp.jsonin the repo root). - Save the file. The server appears in the MCP list; authorize it there if asked.
Authorization
If the service needs sign-in, an authorize button appears next to the server in the MCP list.
Server tool list (15)
Raw names from tools/list. Only developers need these.
| search_memory | Search the user's stored memory for context relevant to a question and return the best matches. Use this before answering anything that depends on what the user told you previously. |
| add_memory | Store a durable fact or preference about the user. Text is run through extraction, so conversational filler is discarded and near-duplicates are merged; a skipped result is normal and not an error. |
| list_memories | List the user's most recent memories without searching. |
| get_memory | Retrieve a single memory by its id, as returned by search or list. |
| delete_memory | Permanently delete one or more memories by id. This cannot be undone, so confirm with the user before calling it. |
| update_memory | Change the labels on an existing memory: tags, importance, metadata, source or event_type. The memory text itself cannot be edited — to correct a fact, call add_memory with the correction instead. |
| delete_all_memories | Delete all memories this connection can reach. Defaults to a dry run that reports the count only; deleting for real requires dry_run=false and the exact confirm string 'DELETE ALL MEMORIES'. Never call this without the user explicitly asking to erase everything. |
| list_entities | Group this user's memories by origin source, extracted category or extracted type, with counts and last activity. Use it to see what is stored before searching, or to find an entity to delete. |
| delete_entity | Delete every memory belonging to one entity returned by list_entities, for example every memory whose source is 'cursor'. Defaults to a dry run; pass dry_run=false to delete after the user confirms. |
| list_memory_events | Read the recent lifecycle log for this user's memory — adds, forgets, summaries and imports, newest first. Use it to explain why a memory is present or missing. |
| get_memory_profile | Return the standing profile derived from this user's memory: communication style, recurring topics, work context and learned preferences. Read this once at the start of a conversation to personalise your replies, instead of inferring it from individual search hits. |
| get_related_memories | Follow the typed relationship graph out from one memory to find what supports, contradicts, summarises, continues or resembles it. Use this when the question is how facts relate, or to check whether a memory you are about to rely on is contradicted by another. Search answers 'what is relevant'; this answers 'what is connected'. |
| get_event_status | Report where one memory is in the ingestion pipeline and whether it is searchable yet. Poll this after add_memory instead of adding again. |
| list_connected_sources | List the knowledge sources connected to this workspace, such as GitHub, Slack, Notion, Google Drive, OneDrive, Granola, S3 or a crawled website, with their sync state. |
| whoami | Report how this client is authenticated and what it is allowed to do. Useful when a write is refused and you need to explain why. |