MCP server for persistent AI memory — give any AI assistant long-term recall. Works with Claude, Kiro, Cursor, Windsurf.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add memory --env VELIXAR_API_KEY='<VELIXAR_API_KEY>' -- npx -y velixar-mcp-server~/.codex/config.toml
[mcp_servers.memory]
command = "npx"
args = ["-y", "velixar-mcp-server"]
env = { VELIXAR_API_KEY = "<VELIXAR_API_KEY>" }
# codex mcp add memory --env VELIXAR_API_KEY='<VELIXAR_API_KEY>' -- npx -y velixar-mcp-server~/.cursor/mcp.json
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"velixar-mcp-server"
],
"env": {
"VELIXAR_API_KEY": "<VELIXAR_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"memory": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"velixar-mcp-server"
],
"env": {
"VELIXAR_API_KEY": "<VELIXAR_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"velixar-mcp-server"
],
"env": {
"VELIXAR_API_KEY": "<VELIXAR_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"velixar-mcp-server"
],
"env": {
"VELIXAR_API_KEY": "<VELIXAR_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"velixar-mcp-server"
],
"env": {
"VELIXAR_API_KEY": "<VELIXAR_API_KEY>"
}
}
}
}Environment variables
VELIXAR_API_KEY
Replace the placeholders with your own credentials and keep them out of version control.