Ratary Memory MCP

Ratary is an AI Brain Platform — infrastructure that gives AI: Persistent memory — durable, owner-scoped, versioned Structured knowledge — metadata…

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

Save Memory: Create a new persistent coding memory entry with title, markdown body, project scope, and optional tags.
Update Memory: Update fields on an existing memory by UUID without creating a new record.
Delete Memory: Permanently delete a memory by UUID. Prefer archive_memory for reversible removal.

What data it sees

Do you need an account

No: the server works without sign-in

Ratary is an AI Brain Platform — infrastructure that gives AI:

Persistent memory — durable, owner-scoped, versioned Structured knowledge — metadata, relations, graph traversal Intelligent retrieval — hybrid search + bounded context assembly Protocol access — Ratary MCP, REST, optional gRPC It sits between AI clients and storage. One brain, many surfaces — Cursor, Claude Code, custom agents, enterprise APIs, and remote MCP hosts.

The runnable deployment is Ratary Server — this repository. Ratary is the product; Ratary Server is what you clone and run.

Bring your model. Ratary brings the memory.

How to connect

How to connect Ratary Memory MCP to Claude

The server address to paste into Claude settings:

https://ontorata.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 Ratary Memory MCP to ChatGPT

The server address to paste into ChatGPT settings:

https://ontorata.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 Ratary Memory MCP to Cursor

The server address to paste into Cursor settings:

https://ontorata.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.json in 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 (28)

Raw names from tools/list. Only developers need these.

save_memory	Create a new persistent coding memory entry with title, markdown body, project scope, and optional tags.
update_memory	Update fields on an existing memory by UUID without creating a new record.
delete_memory	Permanently delete a memory by UUID. Prefer archive_memory for reversible removal.
get_memory	Fetch a single memory by UUID including full content and metadata.
get_memory_by_codename	Fetch a memory by stable codename (e.g. AUTH-0001) when UUID is unknown.
search_memory	Search memories by keyword, tag, or project using hybrid full-text and semantic ranking.
get_memory_by_path	Read a memory by vault source_path when syncing from a file-oriented workflow.
get_context	Rank relevant memories and assemble token-budgeted markdown context for a task.
build_prompt	Build system and user prompts for an external LLM from ranked project memory.
list_projects	List all project slugs that have at least one memory for this owner.
list_tags	List distinct tags used across all memories for this owner.
link_memories	Create a typed relation edge between two memories in the knowledge graph.
list_relations	List graph relations attached to a memory (incoming and outgoing).
toggle_favorite	Toggle the favorite flag on a memory for quick access and ranking.
archive_memory	Soft-archive a memory so it is hidden from default search but recoverable.
get_graph_capabilities	Describe available knowledge-graph traversal modes, limits, and relation types.
traverse_relations	Traverse memory relations via bidirectional BFS from a seed memory or slug.
list_workspaces	List workspaces owned by the authenticated identity.
list_agents	List agent identities registered in the current workspace.
register_agent	Register a new agent identity (Cursor, Claude Code, etc.) in the workspace.
get_capabilities	Return the deployment capability manifest: enabled modules, tool registry, and limits.
negotiate_capabilities	Handshake to verify client-required capability flags against this deployment.
submit_signal	Submit memory quality feedback or forge inspection outcome signals for learning loops.
run_stewardship	Run the memory stewardship pipeline (dedupe, archive candidates). Dry-run by default.
get_compression_status	Report semantic compression status and pending duplicate clusters for the owner.
sync_pull	Pull memory deltas from the server since the last sync cursor for a client platform.
sync_push	Push batched memory create/update/delete operations from a client platform.
sync_status	Get multi-client sync status and last cursor for a platform.