Memory engine for agents with layered decay and promotion, a self-organizing topic tree, hybrid search and LLM-based consolidation.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add engram-rs --env ENGRAM_API_KEY='<ENGRAM_API_KEY>' -- npx -y engram-rs-mcp~/.codex/config.toml
[mcp_servers.engram-rs]
command = "npx"
args = ["-y", "engram-rs-mcp"]
env = { ENGRAM_API_KEY = "<ENGRAM_API_KEY>" }
# codex mcp add engram-rs --env ENGRAM_API_KEY='<ENGRAM_API_KEY>' -- npx -y engram-rs-mcp~/.cursor/mcp.json
{
"mcpServers": {
"engram-rs": {
"command": "npx",
"args": [
"-y",
"engram-rs-mcp"
],
"env": {
"ENGRAM_API_KEY": "<ENGRAM_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"engram-rs": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"engram-rs-mcp"
],
"env": {
"ENGRAM_API_KEY": "<ENGRAM_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"engram-rs": {
"command": "npx",
"args": [
"-y",
"engram-rs-mcp"
],
"env": {
"ENGRAM_API_KEY": "<ENGRAM_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"engram-rs": {
"command": "npx",
"args": [
"-y",
"engram-rs-mcp"
],
"env": {
"ENGRAM_API_KEY": "<ENGRAM_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"engram-rs": {
"command": "npx",
"args": [
"-y",
"engram-rs-mcp"
],
"env": {
"ENGRAM_API_KEY": "<ENGRAM_API_KEY>"
}
}
}
}Environment variables
ENGRAM_API_KEY
Replace the placeholders with your own credentials and keep them out of version control.