Show HN: Sgai – Goal-driven multi-agent software dev (GOAL.md → working code)

Source: Hacker News

Introduction

We built Sgai to experiment with a different model of AI‑assisted development.
Instead of prompting step‑by‑step, you define an outcome in GOAL.md (what should be built, not how), and Sgai runs a coordinated set of AI agents to execute it.

How It Works

• Decomposes the goal into a DAG of roles (developer → reviewer → safety analyst, etc.)
• Asks clarifying questions when needed
• Writes code, runs tests, and iterates
• Uses completion gates (e.g., make test) to determine when the task is actually done

Everything runs locally in your repository. A web dashboard shows real‑time execution of the agent graph, and nothing auto‑pushes to GitHub.

Demo

Demo (4 min)

Repository

• GitHub:
• Open source (Go)
• Works with Anthropic, OpenAI, or local models via opencode

Discussion

We’ve used Sgai internally for prototyping small apps and internal tooling. It’s still early and rough in places, but functional enough to share.

Curious what people think about DAG‑based multi‑agent workflows for coding. Has anyone here experimented with similar approaches?