Graphweave

Visual LangGraph builder where your graph runs exactly as drawn — visually compose nodes, edges, and tools, then stream execution in real time.

Note

Graphweave is in early development. The scaffold is in place but core features are still being built.

What is this?

Every team whiteboard-designs their agent flow, then spends days translating it into LangGraph code — wiring nodes, debugging edges, mapping state. The diagram and the code drift apart immediately.

Graphweave kills that gap. The canvas is the execution. No translation step, no drift. You draw the agent, you run the agent. Same artifact.

The rule

What you draw is what runs. GraphSchema maps 1-to-1 to LangGraph primitives. No abstraction layer between the visual and the execution.

Architecture

packages/
├── canvas/       React 19 + React Flow + Zustand — the visual builder
├── shared/       GraphSchema types — the contract between canvas and execution
├── sdk-core/     Plugin interfaces (transport, events, storage)
└── execution/    FastAPI + LangGraph — builds and runs graphs via SSE

The canvas produces a GraphSchema. The execution layer consumes it, builds a LangGraph StateGraph, and streams node-by-node execution back over SSE.

Node types

Node	Purpose
Start	Entry point — every graph has exactly one
LLM	Calls a language model (Gemini, OpenAI, Anthropic)
Tool	Runs a registered tool with input/output mapping
Condition	Branches on field checks, LLM routing, tool errors, or iteration limits
Human Input	Pauses execution and waits for user input
End	Terminal node — every graph has at least one

Supported providers

• Google Gemini
• OpenAI
• Anthropic

Getting started

Prerequisites

• Node.js >= 20
• pnpm >= 10
• Docker (for the execution layer)
• uv (for Python dependency management)

Setup

# Clone
git clone https://github.com/prosdevlab/graphweave.git
cd graphweave

# Install dependencies
pnpm install

# Copy environment config
cp .env.example packages/execution/.env
# Add at least one LLM provider key to packages/execution/.env

# Start everything (canvas + execution in Docker)
pnpm dev

This runs the React canvas on http://localhost:5173 and the FastAPI execution server on http://localhost:8000.

Verify

# TypeScript + Python checks in parallel
pnpm verify

# Or individually
pnpm typecheck        # TypeScript type checking
pnpm check            # Biome lint + format
pnpm verify:py        # Ruff + pytest

Project structure

Path	What	Managed by
packages/canvas/	React 19 + Vite + Tailwind v4	pnpm
packages/shared/	GraphSchema TypeScript types	pnpm + tsup
packages/sdk-core/	Plugin interfaces	pnpm + tsup
packages/execution/	FastAPI + LangGraph	uv + Docker
docker-compose.yml	Production execution	Docker
docker-compose.dev.yml	Dev execution (hot reload)	Docker

Contributing

Commit conventions

This repo uses Conventional Commits enforced by commitlint.

type(scope): description

# Examples
feat(canvas): add node drag-and-drop
fix(execution): handle missing API key gracefully
chore(deps): bump langchain to 0.3

Types: feat, fix, docs, style, refactor, perf, test, chore, ci, build, revert

Scopes: canvas, shared, sdk-core, execution, docs, deps, docker, schema, skills

Pre-commit hooks

Commits automatically run:

1. .env file guard — blocks accidental secret commits
2. lint-staged — Biome for TS/JS, Ruff for Python
3. typecheck — full tsc --noEmit (cached by Turbo)
4. commitlint — validates commit message format

License

MIT