The server itself exposes 11 tools, and some of them can write to your Mem0 account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Mem0.

How is this different from connecting Mem0 MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Mem0 MCP API changes?

We track the Mem0 MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Mem0 tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Mem0?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Mem0 MCP server myself?

No. TrustGate connects to the Mem0 MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Mem0 directly.

Connect Mem0 MCP to your AI agents, securely.

One secured endpoint for every Mem0 call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Mem0 MCP now

What your agents can ask once Mem0 MCP is connected.

Select the tools your AI agents can call

11 tools from the Mem0 MCP schema.

Showing 11 tools

search_memoriesreadSemantic search across existing memories, with filters.
get_memoriesreadLists memories with structured filters and pagination.
get_memoryreadRetrieves one memory by its memory_id.
add_memorywriteSaves text or conversation history for a user or agent. What is written here is read back as trusted context in later sessions.
update_memoryhigh riskOverwrites a memory's text and/or metadata after confirming the ID. It replaces rather than merges.
delete_memoryhigh riskDeletes a single memory by its memory_id.
delete_all_memorieshigh riskBulk-deletes all memories in scope, with no narrower target than the scope itself.
list_entitiesreadEnumerates the users, agents, apps and runs stored in Mem0.
delete_entitieshigh riskDeletes a user, agent, app or run entity along with all of its memories.
list_eventsreadLists memory operation events with filters and pagination.
get_event_statusreadChecks the status of an asynchronous memory operation by its event_id.

What Mem0 MCP exposes

Mem0 MCP groups its 11 tools into 3 categories, each with a suggested starting policy.

All 11 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Memories	Saving text or conversation history for a user or agent, semantic search across existing memories with filters, listing them with structured filters and pagination, retrieving one by its ID, overwriting a memory's text and metadata, deleting a single memory, and bulk-deleting everything in scope	Read / Write	delete_all_memories has no narrower target than the current scope, so approve it by hand every time. update_memory overwrites rather than merges, so a partial write loses whatever it did not restate
Entities	Enumerating the users, agents, apps and runs stored in Mem0, and deleting one of those entities along with all its memories	Read / Write	delete_entities removes a whole identity and everything stored against it in one call, which makes it broader than any per-memory delete. list_entities is the safe way to see the shape of the account first
Events	Listing memory operation events with filters and pagination, and checking the status of an asynchronous memory operation by its event ID	Read	Read-only and the closest thing to an audit trail on this connection. Worth allowing freely, since it is how you confirm what an agent actually wrote

Connect Mem0 MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Mem0
Find Mem0 in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Mem0 MCP or block it. You allow the exact tools each agent can call, one by one.

Mem0 MCP	11 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 11 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Mem0 MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Mem0 MCP11 documented tools on a hosted server at mcp.mem0.ai/mcp. Nothing runs locally and memories live in the Mem0 account rather than on the machine. Browser sign-in or an API key as a bearer token; without either the server returns 401.
Coding assistants	Connection & credentialsper user	Memories7 tools
Internal copilots	Tool access policiesdeny by default	Entities2 tools
Workflow automations	Request & response inspectioninline	Events2 tools
Compatible MCP clients	Activity monitoringalerts and logs