Memory

What it does

A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations. It is the simplest way to give an agent recall across sessions without standing up a vector database. Expect to manage the file yourself; there is no built-in compaction or expiry.

Install

npx -y @modelcontextprotocol/server-memory

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-memory"
      ]
    }
  }
}

When to use it

Reach for this when the task needs memory inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

• Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
• Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude
• Everything — A deliberately exhaustive reference server that exercises prompts, resources, tools, sampling, and notifications