SmritiTantra

Memory for AI agents: ingest, recall, learn — one shared async core, served three ways: embedded in your process, over HTTP, or as an MCP server.

Full documentation →

Install

A bare install is a lightweight remote SDK (smrititantra.connect(), models, exceptions, the sync facade):

pip install smrititantra

Everything heavier lives behind an extra — mix in what you need:

pip install "smrititantra[local]"    # embedded engine + Postgres store
pip install "smrititantra[server]"   # HTTP + MCP service
pip install "smrititantra[all]"      # local + server + every scale backend

Quickstart

import asyncio
from smrititantra import Smrititantra

async def main() -> None:
    async with Smrititantra() as smriti:
        await smriti.concepts.author(
            kind="metric", name="net_revenue",
            metadata={"title": "Net Revenue", "description": "Gross minus refunds."},
            body="Net revenue is gross order value minus refunds.",
        )
        await smriti.snapshots.compile()
        pack = await smriti.recall("what is net revenue")
        print(pack.confidence, [i.ref for i in pack.required])

asyncio.run(main())

Or try the self-contained demo (no database or config needed):

pip install "smrititantra[local]"
smrititantra demo

CLI

smrititantra --help covers the full surface: orgs, sources, sync, import, mine-workload, mine, coverage, gaps, reviews, compile, recall, search, get, graph, snapshots, config, agents, keys, serve, mcp, demo, init. Details in the CLI reference.

License

Apache-2.0 — see LICENSE.