---
title: "Zedflow, agent flows you can read, compose and control"
description: "My agent flow engine: every flow is a typed graph, validated before it runs, with human approvals checked on typed data and a trace of every turn."
url: "https://mc-stud.io/en/work/zedflow-agent-flows"
author: "Matisse Chenavas"
date: "2026-10-09"
lang: "en"
ai: "true"
---

# Zedflow, agent flows you can read, compose and control

> With Zedflow, an agent flow is read, validated and composed like a program, and sensitive decisions wait for a human approval.

- **Client**: MC Studio, research and development
- **Sector**: AI agent tooling
- **Period**: 2026
- **My role**: Designer and sole developer
- **Setting**: Internal project
- **Team**: Solo · 1 person
- **Status**: Pilot
- **Technologies**: Rust, TypeScript, Vue.js, ADK-Rust
- **Reading time**: 2 min

## At a glance

- **Stake**: Handing agents tasks that commit the business requires reading what they will do, controlling what they see and reviewing what they did.
- **Mechanism**: A strict graph, validated before it runs, where a context strategy composes what each model receives and every turn is traced.
- **Proof**: The engine, its web studio and the zf command-line tool, published on GitHub and used to prepare my monthly invoicing.
- **Impact**: An agent becomes auditable: you know what it saw, what it did and where a human decided.
- **Limits**: Zedflow is a preview with no external user; part of the execution is designed but not implemented yet.

## The starting point

AI agents can chain actions. What is most often missing is the ability to read what they will do before launching them, to control what they see, and to find out afterwards why they acted as they did. Without that, it is hard to hand them anything but tasks without consequences.

## What I built

Zedflow is an agent flow engine and studio[^adk]. A flow is a typed graph, checked before it runs: an invalid transition or a badly declared value is refused with a precise diagnostic, instead of failing in the middle of a run.

At every turn, a context strategy decides, block by block, what the model receives; state lives in typed stores, not in the prompt. Bridges connect flows to each other, each with its own responsibility, context and tools. Human approvals are steps of the graph, checked on typed data: an agent can prepare an invoice, it cannot send it until the person in charge has approved.

> [!NOTE] A software factory, not a single agent
> The guiding idea: one specialised flow per responsibility, composed with the others like a production line, with humans approving what matters.

## What the evidence shows

The code is public, with the engine, the web studio and the command-line tool[^git]. You can try it locally with a demo provider, without an account or a real model. Every turn can be reviewed in the studio: what the model received, which tools ran, what changed.


## What it changes

For a client, this approach makes an agent auditable, and that is the condition for handing it tasks that commit the business. I apply it to myself first: Zedflow prepares my monthly invoicing, and nothing leaves without my approval[^billing]. It is also a testing ground for what I recommend to my clients: useful, bounded agents whose every decision can be reviewed.

## Limits

- Formats and interfaces still change, and part of the execution is designed but not implemented yet.
- No external user to date; the only real use is internal to MC Studio.
- The licence, PolyForm Noncommercial 1.0.0, reserves commercial use for a separate agreement.

[^adk]: Zedflow builds on ADK-Rust, whose graph workflows follow the same principles as LangGraph.

[^git]: 690 commits, all mine, from 8 July 2026 to 3 October 2026 (default branch of the public repository, as of 12 October 2026).

[^billing]: Agents prepare the invoice and email drafts; sending waits for my explicit approval, a step of the graph.

## Links

- [GitHub repository](https://github.com/ZEDIUM-Off/zedflow)

---

Matisse Chenavas, Founder of MC Studio. https://mc-stud.io/en/work/zedflow-agent-flows