Skip to content

Memory

Persistent key-value storage and knowledge graph for maintaining context across sessions.

  • Store entities - Save named entities with observations
  • Create relations - Link entities together
  • Query knowledge - Retrieve stored information
  • Persistent storage - Data persists between sessions
  • Useful for remembering user preferences, project context
  • Entities have names and lists of observations
  • Relations connect entities (e.g., “project X uses technology Y”)
  • Storage is local to the machine

Required config.json:

{
"name": "Memory",
"slug": "memory",
"enabled": true,
"provider": "memory",
"type": "mcp",
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-memory"]
}
}
  • What kind of information do you want to persist? (preferences, project notes, etc.)
  • Should this be shared across workspaces or specific to one?
{
"allowedMcpPatterns": [
{ "pattern": "read", "comment": "Read stored entities" },
{ "pattern": "search", "comment": "Search knowledge graph" },
{ "pattern": "open", "comment": "Open knowledge graph nodes" }
]
}

This is a local server - no API keys or OAuth needed.