Ariana is an open-source, agentic development platform aiming to be an alternative to tools like Cursor and Devin. It empowers developers by running parallel AI agents that automate coding, GitHub management, environment setup, testing, and deployment, leveraging Anthropic's Claude models and hosted on Hetzner VPSs. The platform offers a cloud version and aims to streamline complex development workflows.

How It Works

The core architecture features parallel, autonomous coding agents operating on remote Hetzner VPSs. These agents utilize the Anthropic SDK to perform tasks such as code editing, GitHub issue management, automated branching, committing, and pushing. Agents can spawn sub-agents, manage environments, and even host services with auto-generated SSL certificates, enabling a distributed and automated development process.

Quick Start & Requirements

Setup is currently described as "unpractical" due to significant third-party service and custom infrastructure dependencies. A detailed checklist of prerequisites and setup steps is marked as Work In Progress (WIP). Users are directed to community channels for current setup guidance.

Highlighted Details

• Supports parallel coding agents for background task automation.
• Features automated GitHub workflows: issue integration, branch creation/naming, auto-commit, and auto-push.
• Agents can manage development environments, run Docker Compose, and host public services with SSL.
• Includes agent forking for state persistence and gaming-grade desktop streaming for remote interaction.
• Offers markdown/mermaid rendering and advanced diff views via diffs.com.
• Provides a mobile web version for on-the-go development access.

Maintenance & Community

The project is actively developing a "big V2" and encourages community engagement via Discord and X. Links to these platforms are provided within the README.

Licensing & Compatibility

Licensed under AGPL-3.0. This strong copyleft license requires derivative works to also be open-sourced under the same license, potentially posing compatibility challenges for integration into closed-source commercial products.

Limitations & Caveats

The primary limitation is the current difficulty and impracticality of local setup, requiring substantial custom infrastructure. The project is heavily reliant on third-party services and Anthropic's API, which may incur costs or introduce external dependencies.