MCP AutoMem — Claude Code MCP server: memory and setup
Connects Claude Code to persistent memory so it can retrieve project context across sessions. Requires a running AutoMem service.
claude mcp add mcp-automem -- npx -y @verygoodplugins/mcp-automemWhat MCP AutoMem does
MCP AutoMem connects Claude Code to the AutoMem memory service. The assistant can save useful context and retrieve it in later sessions. Storage uses a graph-vector approach backed by FalkorDB and Qdrant, and connected MCP clients can share the same memory.
Start the AutoMem service locally or deploy it separately, then connect the MCP server. The Claude Code plugin bundles the server with hooks for background recall and memory-store tracking. Railway deployments require an API key; the local setup can run without one. The MCP server needs a running AutoMem service to connect to.
The guided installer, npx @verygoodplugins/mcp-automem install, displays a review plan before writing changes and keeps .bak copies of modified files. For a local Claude Code connection, run claude mcp add mcp-automem -- npx -y @verygoodplugins/mcp-automem; a remote service can use HTTP transport instead.
What it helps with
- carry project decisions into later sessions
- retrieve coding preferences shared earlier
- share stored context across MCP clients
- bring back details from a previous conversation
Example prompts for the agent
- Remember that this project uses PostgreSQL for its database.
- What coding conventions did we settle on earlier?
- Save this architecture decision for future sessions.
How to install MCP AutoMem
1Platform CLI
claude mcp add mcp-automem -- npx -y @verygoodplugins/mcp-automem2Platform CLI: remote server
claude mcp add --transport http mcp-automem https://{endpoint}/mcp1Platform CLI
codex mcp add mcp-automem -- npx -y @verygoodplugins/mcp-automem2Platform CLI: remote server
codex mcp add mcp-automem --url https://{endpoint}/mcp1Config file
{
"mcpServers": {
"mcp-automem": {
"command": "npx",
"args": [
"-y",
"@verygoodplugins/mcp-automem"
]
}
}
}2Config file
{
"mcpServers": {
"mcp-automem": {
"url": "https://{endpoint}/mcp"
}
}
}1Config file
{
"servers": {
"mcp-automem": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@verygoodplugins/mcp-automem"
]
}
}
}2Config file
{
"servers": {
"mcp-automem": {
"type": "http",
"url": "https://{endpoint}/mcp"
}
}
}1Platform CLI
gemini mcp add mcp-automem npx -y @verygoodplugins/mcp-automem2Platform CLI: remote server
gemini mcp add -t http mcp-automem https://{endpoint}/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about MCP AutoMem
How do I install MCP AutoMem in Claude Code?
For a local connection, run claude mcp add mcp-automem -- npx -y @verygoodplugins/mcp-automem. Start the AutoMem service first.
What does MCP AutoMem do?
It links an AI assistant to a memory service, letting it save context and retrieve it in later sessions. The storage system uses a graph-vector approach.
Does MCP AutoMem need an API key?
A local service can be used without an API key. A Railway deployment requires one.
