SMRITI Memory MCP
A neuro-inspired long-term memory engine for AI agents that provides persistent, adaptive recall via a dual-process architecture. It features System 1 for immediate, non-blocking ingestion and System 2 for asynchronous consolidation into semantic graphs. The server supports privacy-first private rooms, persona palaces, and native integration with MCP clients like Claude Code and Gemini.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Provide persistent long-term memory for AI agents
- Consolidate raw interactions into semantic knowledge graphs
- Isolate user memories in private rooms for privacy
- Integrate with MCP clients like Claude Code
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
smriti-memcore 1.4.17
{
"version": "1.4.17",
"transport": "stdio",
"identifier": "smriti-memcore",
"registryType": "pypi"
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"smriti-memory-mcp": {
"command": "uvx",
"args": [
"smriti-memcore==1.4.17"
]
}
}
}Cline
{
"mcpServers": {
"smriti-memory-mcp": {
"command": "uvx",
"args": [
"smriti-memcore==1.4.17"
]
}
}
}VS Code
{
"servers": {
"smriti-memory-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"smriti-memcore==1.4.17"
]
}
}
}Claude Desktop
{
"mcpServers": {
"smriti-memory-mcp": {
"command": "uvx",
"args": [
"smriti-memcore==1.4.17"
]
}
}
}Let your agent find its tools
Find servers like SMRITI Memory MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.