Windmill — Open-Source Internal Tool Platform

What it is

Windmill is an open-source platform for building internal tools, APIs, background jobs, and workflows. You write scripts in Python, TypeScript, Go, Bash, or SQL, and Windmill generates a UI, handles scheduling, manages secrets, and provides approval flows -- all without frontend code.

It targets engineering and operations teams that need to build internal tooling quickly without maintaining a separate web application. Windmill is self-hostable and AGPLv3 licensed.

How it saves time or tokens

Building an internal tool traditionally means writing a script, then building a web UI for non-technical users to trigger it, then adding authentication, secret management, and scheduling. Windmill collapses all of this into a single platform. Write a function, Windmill infers the input schema and generates a form UI.

The platform claims to be 13x faster than Airflow for workflow orchestration, with a focus on developer experience over configuration complexity.

How to use

1. Deploy Windmill with Docker Compose:

curl https://raw.githubusercontent.com/windmill-labs/windmill/main/docker-compose.yml -o docker-compose.yml
docker compose up -d

1. Open http://localhost:8000 and create your first script.

1. Write a script in any supported language:

# Windmill auto-generates a UI from this function signature
def main(name: str, count: int = 5):
    return [f'Hello {name}' for _ in range(count)]

Windmill creates a web form with a text input for name and a number input for count.

Example

// TypeScript script in Windmill
import { Windmill } from 'windmill-client';

export async function main(
  database_url: string,
  query: string
): Promise<Record<string, unknown>[]> {
  const client = new Windmill();
  // Windmill manages the database_url as a secret
  const results = await client.runQuery(database_url, query);
  return results;
}

Workflows chain multiple scripts together with branching and error handling:

fetch_data -> transform -> if error: notify_slack else: write_to_db

Related on TokRepo

• AI Tools for Automation -- More workflow automation platforms
• AI Tools for No-Code -- Low-code and no-code development tools

Common pitfalls

• The AGPLv3 license requires you to open-source modifications if you offer Windmill as a service to third parties. For internal use only, this does not apply.
• Self-hosted Windmill needs PostgreSQL as its backing database. The Docker Compose file includes Postgres, but production deployments should use a managed database.
• Script dependencies are installed at runtime. Cold starts can be slow for scripts with heavy dependencies. Use Windmill's dependency caching to mitigate this.