# Hanzo Flow — Hanzo AI

> Build AI agents, RAG pipelines, and multi-step workflows with a drag-and-drop canvas.

flow.hanzo.ai

# Hanzo Flow

Build an AI workflow on a canvas, ship it as an API

Drag a model, a vector store and a prompt onto a canvas, wire them together, and run it with the values you actually have. Every component is Python you can open and edit in place, so the canvas stops where your code starts instead of at a wall. When it works, publish it: a REST endpoint, a JSON file a Python app loads, or an MCP server, which turns the flow into a tool any MCP client can call.

[Open Flow](https://flow.hanzo.ai)[Documentation](https://docs.hanzo.ai/docs/cli/flow)

Paste and ship

Paste this into any agent. It reads the skill manifest and calls Flow from there.

Agent promptCopy

```
Read https://hanzo.ai/skill.md and use Hanzo Flow in my project. Start with: hanzo flow workflows list
```

## The canvas is not the ceiling

Visual builders usually work right up until you need the thing they did not think of. This one hands you the source.

01

### The library

Models, vector stores, document loaders, text splitters, output parsers, memory, retrievers and toolkits — plus anything you write yourself, which is a component like the rest.

02

### Open any node

A component is a Python class. Edit it in the browser and the change is live on the next run — no rebuild, no plugin format, no waiting for someone to add the parameter you need.

03

### Step through it

The playground runs a flow one node at a time and shows what each one received and returned. When a chain gives a strange answer, this is where you find the node that caused it.

04

### Multi-agent

Several agents on the same canvas, with conversation state between them and retrieval underneath — visible as a graph rather than buried in a prompt.

05

### Publish it three ways

As a REST endpoint, as JSON your Python app loads directly, or as an MCP server — after which the flow is a tool your coding agent can call.

06

### Keys stay out of the flow

Credentials live as variables the graph references by name, so a flow you export or share carries the wiring and not the secrets.

Python

Every component

REST

Or JSON, or MCP

:7860

On your machine

MIT

Self-hosted

Open source royalty

### 20% to the author of the code that runs

Verify that you own a repository, and every org that deploys a project built from it pays you 20% of its metered spend, accrued each period.

[How authors are paid](https://hanzo.ai/oss)[Our repositories](https://hanzo.ai/open-source)

## Run it locally first

uv pip install hanzoflow, then uv run hanzoflow run. It opens on 127.0.0.1:7860. Docker works too.

[Open Flow](https://flow.hanzo.ai)[View on GitHub](https://github.com/hanzoai)

## Start building with Flow

[Get your API key](https://platform.hanzo.ai/api-keys)

One API key. One credit balance. Every primitive.

## Open source

License: Apache-2.0[hanzoai](https://github.com/hanzoai)

## Get Flow

Visual workflow builder

[Deploy to Cloud](https://platform.hanzo.ai/deploy)