graph-ted-db

A graph database in a folder. No server, shares like files.

pip install graph-ted-db
from graph_ted_db import GraphStore, init_graph
init_graph("./my-graph", name="demo", exist_ok=True)
g = GraphStore.open("./my-graph")
g.make_node(labels=["Person"], props={"name": "Alice"})

Requires Python 3.10+. Until the package is on PyPI, install from a git checkout: pip install -e .

Principles

  • Local-first

    A graph is a folder of plain files on your own machine. There is no server to run for the default path, and your work does not depend on a connection. Open it in-process from Python or serve it over HTTP. Multiple devices can add and edit at the same time; concurrent edits to the same record resolve to the latest version. Tested end to end with OneDrive via rclone; Dropbox and other sync services are expected to work.

    Sharing a graph: read this first

  • Your storage and network are the security boundary

    Data is stored as ordinary files. Protect a graph the way you protect your other files: disk encryption, account permissions, backups, and what you expose on the network.

  • Open and agent-friendly

    Readable on-disk formats, a documented Python API, and docs published as llms.txt so agents can read them too.