Cerebro MCP Server

Cerebro MCP server for ZCode — 17 memory tools. Auto-reads ~/.config/cerebro/config.json. Use via: npx -y @mingxy/cerebro-mcp-zcode

Remote serverstreamable-http

What is the Cerebro MCP server?

Cerebro MCP server for ZCode — 17 memory tools. Auto-reads ~/.config/cerebro/config.json. Use via: npx -y @mingxy/cerebro-mcp-zcode. That is what the cerebro mcp server brings to an AI assistant: the same capability, reachable through the Model Context Protocol rather than a separate app or dashboard.

The short version

Cerebro gives AI agents shared persistent memory — across sessions, devices, agents, and teams. One API key reconnects everything. Self-hosted, open-source.

The tools it exposes

The server publishes 14 tools. What each one is for:

  • Concept — What it is
  • memory_store — Save facts, decisions, preferences with smart dedup
  • memory_search — 11-stage hybrid search (vector + BM25 + reranker)
  • memory_get — Retrieve a specific memory by ID
  • memory_update — Modify content, tags, importance, tier
  • memory_delete — Remove a memory
  • memory_list — Browse memories with filters and pagination
  • memory_ingest — Smart-ingest full conversations (LLM extraction)
  • memory_profile — Auto-generated user profile (static facts + dynamic context)
  • memory_stats — Analytics: tag distribution, decay curves, relation graphs
  • space_create — Create Team or Organization spaces
  • space_list — List all accessible spaces

What it needs from you

Configuration is passed through the environment: OMEM_API_URL, OMEM_API_KEY, OMEM_EMBED_API_KEY, AWS_ENDPOINT_URL, OSS_ACCESS_KEY_ID, OSS_ACCESS_KEY_SECRET. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.

Getting it running

Because this one is hosted, setup is mostly authentication — you point your client at the endpoint and approve access. Nothing runs on your machine, so there is no runtime to keep patched.

How it compares

This sits in the knowledge and memory group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Cerebro's toolset — Concept, memory_store, memory_search and 11 more — is a fair guide to whether it matches your workflow. It is maintained by mingxy; worth a glance at recent repository activity before you build anything load-bearing on it.

We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.

Things to watch

  • Your data travels to the provider's service, so the usual questions apply about what you send and what they retain.
  • With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Cerebro.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

Available tools

ToolWhat it does
ConceptWhat it is
memory_storeSave facts, decisions, preferences with smart dedup
memory_search11-stage hybrid search (vector + BM25 + reranker)
memory_getRetrieve a specific memory by ID
memory_updateModify content, tags, importance, tier
memory_deleteRemove a memory
memory_listBrowse memories with filters and pagination
memory_ingestSmart-ingest full conversations (LLM extraction)
memory_profileAuto-generated user profile (static facts + dynamic context)
memory_statsAnalytics: tag distribution, decay curves, relation graphs
space_createCreate Team or Organization spaces
space_listList all accessible spaces
space_add_memberInvite users to a shared space
memory_shareShare a memory to any space with provenance

How to install the Cerebro MCP server

{
  "mcpServers": {
    "cerebro": {
      "command": "npx",
      "args": ["-y", "skills"],
      "env": {
        "OMEM_API_URL": "your-value",
        "OMEM_API_KEY": "your-value",
        "OMEM_EMBED_API_KEY": "your-value",
        "AWS_ENDPOINT_URL": "your-value",
        "OSS_ACCESS_KEY_ID": "your-value",
        "OSS_ACCESS_KEY_SECRET": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

VariableDescriptionRequired
OMEM_API_URLEndpoint or connection string the server talks to.Yes
OMEM_API_KEYCredential the server authenticates with.Yes
OMEM_EMBED_API_KEYCredential the server authenticates with.Yes
AWS_ENDPOINT_URLEndpoint or connection string the server talks to.Yes
OSS_ACCESS_KEY_IDCredential the server authenticates with.Yes
OSS_ACCESS_KEY_SECRETCredential the server authenticates with.Yes

Example prompts to try

  • Use Cerebro to Concept.
  • Use Cerebro to memory store.
  • Use Cerebro to memory search.

Frequently asked questions

It connects Cerebro to MCP-compatible AI assistants such as Claude and Cursor, exposing 14 tools (Concept, memory_store, memory_search, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Cerebro directly.