[Open Source]

Typed graph engine,
versions like git

Schema-as-code. Branch and merge. Local paths, S3, or HTTP.

macos / linux

schema.pg

Builtto Branch.

[Branching]

Branch your data like you branch code.

Every agent gets an isolated branch. Enrich, migrate, or remodel in parallel without touching main. Review the results and merge what works.

[Branch Graph]

[Isolation]

Isolated writes, clean reviews

Each agent writes to its own copy-on-write branch. Main stays untouched while schema validation runs on every mutation.

[Collaboration]

Humans and agents on one graph

A human starts a branch, an agent picks it up. Or an agent enriches data and a human reviews it. The branch is the handoff.

[Experimentation]

Try it on a branch first

Schema migrations, bulk enrichments, new edge types. Test any change on a branch. If it breaks, discard. Main stays clean.

[Shaping]

Define the shape. The graph follows.

[Features]

One engine, the whole graph layer.

Git Workflows

Branches, merges, transactional runs.Git-style graph workflows. On your data, not your code.

Branch Queue4 active

in progress · 2

research/hybrid-ranking4 commits

research/claim-graph3 commits

merged · 1

feat-hybrid → mainv0.5.0 · 21ms

Schema-as-code

validation + linting at compile time.

$omnigraph schema plan ./repo

✓ 4 types compiled

✓ 11 of 11 queries typechecked

✓ lint clean

Single CLI

init · load · read · change · branch · merge · snapshot

$init$load$read$change$branch$merge$snapshot

$load$read$change$branch$merge$snapshot$init

$read$change$branch$merge$snapshot$init$load

Typed Schema

node Paper {
  doi: String @key
  title: String
  emb: Vector(768) @embed
}
edge Cites: Paper → Paper
edge Supports: Paper → Claim

Search + Traversal

traversal · bm25 · vector · rrf · fuzzy · text

paper.title

traversalbm25vectorrrffuzzytext

Paper.cites → Paper.embedding3 hops

Policy-as-code

Cedar for server-side access.

S3-native Storage

AWS S3 · RustFS · ON-prem · hybrid · VPC

Typed schema, typed queries, typed mutations. No silent drift. No query surprises.

[Use Cases]

Model any domain in a typed graph

Research Agents

Papers, notes, concepts, and claims connected in a typed graph, giving agents structured context for citation tracing and synthesis.

Context Graphs

On-prem agent memory that survives sessions. Typed schema per agent, branches for experiments, server-side Cedar policies.

Revenue Operations

Accounts, contacts, deals, signals, and activities in one typed graph, so agents can query risk, gaps, and account context.

Incident Response

Services, alerts, incidents, and actions in a dependency graph for blast-radius queries, past-incident lookup, and response context.

Compliance

Policies, controls, evidence, and risks versioned in a graph for gap checks, decision traces, and review workflows.

Enterprise Knowledge

Policy-gated knowledge for the whole org. Per-team branches, server-side Cedar access control, and auditable graph history.

Industry Intel

Companies, people, deals, filings, and patents connected over time for signal detection, actor mapping, and analyst briefings.

[Foundation]

Open source all the way down.

Memory safety without a garbage collector. Release binaries for the CLI and server, with minimal runtime dependencies.

Columnar memory layout for efficient execution and interchange across analytical systems.

Open format with built-in versioning. Each write is an immutable snapshot. Random access, not just column scans.

Production-grade query planner. Compiles typed graph queries to optimized execution plans.

S3-compatible storage for local, on-prem, and cloud deployments with the same repo abstraction.

Amazon's authorization language for server-side, auditable access control.

[Get Started]

Bootstrap Omnigraph
in 60 seconds.

Omnigraph is open source, ships release binaries, and is ready to run.

GitHub Read the Docs

Typed graph engine, versions like git