Smriti
Allows syncing memories to GitHub for backup and sharing via git-based synchronization.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Smriticapture that we chose Go for the new API service"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Smriti
Local-first persistent memory for AI agents via MCP.
Smriti (स्मृति) — Sanskrit for "memory, remembrance"
One brain. Every agent. Zero cloud. Zero cost.
What is this?
A standalone MCP server backed by sqlite-vec and a local embedding model. Install it, point any MCP-compatible agent at it, and every AI you use shares one persistent, semantically searchable memory.
Smriti | Cloud alternatives | |
Setup |
| Accounts + API keys + config |
Cost | $0 | Variable |
Privacy | 100% local | Data on external servers |
Offline | Full functionality | Needs internet |
Portability | Single | DB export/migration |
Related MCP server: OmniHub
Install
npm install -g smritiUsage
# MCP server — stdio mode (for Claude Code, Cursor, etc.)
smriti
# MCP server — HTTP mode (for remote agents)
smriti --http --port 3838CLI Quick Reference
# Capture a memory
smriti capture "We chose PostgreSQL over MySQL for the new service"
# Semantic search
smriti search "database decisions"
# Browse recent memories (last 7 days)
smriti recall --days 7
# Browse by topic
smriti recall "authentication"
# Batch-extract memories from a conversation or document
smriti ingest --text "Long meeting notes or conversation log..." --threshold 0.4
# Memory stats
smriti stats
# Sync to GitHub (requires: smriti auth)
smriti sync
# GitHub auth
smriti auth
smriti whoami
smriti logoutMCP Client Configuration
Claude Code
Add to ~/.claude/mcp.json:
{
"mcpServers": {
"memory": {
"command": "smriti"
}
}
}Cursor
Add to MCP settings:
{
"mcpServers": {
"memory": {
"command": "npx",
"args": ["smriti"]
}
}
}Any MCP client (HTTP mode)
smriti --http --port 3838Then point your client at http://localhost:3838/mcp.
Tools
Tool | Description |
| Store a thought with auto-extracted metadata |
| Semantic search — find thoughts by meaning |
| Browse recent memories with filters |
| Delete a specific memory by ID |
| Get structured context bundle for a topic |
| Memory patterns and insights |
Resources
URI | Description |
| Last 24h of thoughts |
| Topic index with counts |
| People mentioned + context |
| Overall memory statistics |
Prompts
Name | Description |
| Guided capture session |
| End-of-week synthesis |
| Import memories from other sources |
Configuration
Config lives at ~/.smriti/config.json:
{
"db_path": "~/.smriti/brain.db",
"embedding": {
"provider": "onnx",
"model": "Xenova/all-MiniLM-L6-v2"
},
"extraction": {
"provider": "rules"
},
"server": {
"transport": "stdio",
"port": 3838
}
}How it works
You (or an agent) call
capturewith textSmriti generates a vector embedding locally (all-MiniLM-L6-v2 via ONNX)
Regex-based extraction pulls out people, topics, actions, and classifies the type
Everything is stored in a single SQLite file with sqlite-vec for vector search
searchfinds thoughts by semantic similarity, not just keywordsAll data stays on your machine — nothing leaves localhost
License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/moltizmo/smriti'
If you have feedback or need assistance with the MCP directory API, please join our Discord server