Turns files and conversations into memory in local vaults that connected models can search; requires the Mnemosyne OS desktop app.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add mnemosyne-neural-os -- npx -y @mnemosyne_os/mcp~/.codex/config.toml
[mcp_servers.mnemosyne-neural-os]
command = "npx"
args = ["-y", "@mnemosyne_os/mcp"]
# codex mcp add mnemosyne-neural-os -- npx -y @mnemosyne_os/mcp~/.cursor/mcp.json
{
"mcpServers": {
"mnemosyne-neural-os": {
"command": "npx",
"args": [
"-y",
"@mnemosyne_os/mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"mnemosyne-neural-os": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@mnemosyne_os/mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"mnemosyne-neural-os": {
"command": "npx",
"args": [
"-y",
"@mnemosyne_os/mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"mnemosyne-neural-os": {
"command": "npx",
"args": [
"-y",
"@mnemosyne_os/mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"mnemosyne-neural-os": {
"command": "npx",
"args": [
"-y",
"@mnemosyne_os/mcp"
]
}
}
}