Show HN: Multi-agent orchestration layer for experimentation

Category: infrastructure

Tags: agent-orchestration, docker, multi-agent

Score: 5.3/10 (Innovation: 5, Technical: 6, Documentation: 6, Utility: 4)

Chassis is a minimal Docker-based orchestration layer for managing fleets of AI agents using cron-driven tasks and a privileged tool dispatcher. It provides an interesting abstraction around agents as directories with system prompts and configs, with a multi-tenant branch-based workflow for experimentation. The project is early-stage with limited adoption but shows a thoughtful approach to agent security and isolation.

Target audience: backend devs, data engineers

Repository: https://github.com/theo-kirby/chassis · Python · MIT

View on Hacker News