Skip to content

Quickstart

Source mirrored from seocho/QUICKSTART.md

This is the shortest path to a working ontology-aligned graph memory.

You will:

  1. define a tiny ontology
  2. add one sentence
  3. ask a question against the graph memory
Terminal window
uv pip install "seocho[local]"

seocho[local] includes the local SDK engine, agent dependencies, and the embedded LadybugDB graph path. You do not need to run a server for this first example.

from seocho import Seocho, Ontology, NodeDef, RelDef, Property
ontology = Ontology(
name="work",
nodes={
"Person": NodeDef(properties={"name": Property(str, unique=True)}),
"Company": NodeDef(properties={"name": Property(str, unique=True)}),
},
relationships={
"WORKS_AT": RelDef(source="Person", target="Company"),
},
)
client = Seocho.local(ontology)
client.add("Marie Curie worked at the University of Paris.")
print(client.ask("Where did Marie Curie work?"))

What happened:

  • the ontology declared the allowed graph shape
  • add() extracted graph facts that fit that shape
  • ask() queried the graph memory and produced an ontology-grounded answer

The finance-compliance example is the fastest complete project-shaped path:

Terminal window
export OPENAI_API_KEY=...
python examples/finance-compliance/quickstart.py

It ships:

  • examples/finance-compliance/ontology.py
  • six short mock compliance documents
  • a script that ingests them and asks cross-document questions

If a SEOCHO runtime is already running:

from seocho import Seocho
client = Seocho.remote("http://localhost:8001")
print(client.ask("What do we know about ACME?"))

To start the local UI/API/DozerDB stack from a cloned repository:

Terminal window
make setup-env
make up

Then open:

  • UI: http://localhost:8501
  • API docs: http://localhost:8001/docs
  • DozerDB browser: http://localhost:7474
NeedStart here
Understand the projectREADME.md
Use your own ontology and filesdocs/APPLY_YOUR_DATA.md
Learn the Python SDKdocs/PYTHON_INTERFACE_QUICKSTART.md
Run the full platformdocs/RUNTIME_DEPLOYMENT.md
See generated files and tracesdocs/FILES_AND_ARTIFACTS.md