#649 · Primary category: AI Agents & Automation

Shannon

agent ai multi-agent-systems

A production-oriented multi-agent orchestration framework.

Project last updated:08/08/26

GitHub Stars

2.2K

Forks

351

Contributors

7

License

MIT

Why we included this project

Shannon targets teams that have moved past agent demos and now need a runtime they can actually operate. It ships as a self-hostable stack: a Go gateway, Temporal-backed orchestrator, Rust agent core, and Python LLM service in Docker containers with a one-command install. The operational layer is what stands out: hard per-task token budgets with automatic model fallback, real-time event streaming, Prometheus metrics, OpenTelemetry tracing, OPA approval gates, and time-travel debugging that replays any workflow step. It also routes tasks by complexity across strategies like DAG, ReAct, research, or swarm, and exposes an OpenAI-compatible API so existing clients can connect without changes.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category