Orbit v0 launches as a framework for multi-agent safety and security evals
ghadfield · x · 2026-07-23
Orbit v0 is a new framework for multi-agent safety and security research.
- It sits on top of Inspect and is designed to cut down the weeks of infrastructure work that usually come before the first experiment result.
- Researchers can swap out environments, threat models, and agent topologies without rebuilding the whole stack.
- The release is aimed at people running multi-agent experiments in safety/security settings, where fast iteration and controlled comparisons matter.
Related event: Orbit v0: A New Framework for Multi-Agent Safety Eval(2 posts)→
More from coding & agent
- The browser main thread is expensive: a practical guide to JavaScript and CSS animation cost — jh3yy · 2026-09-11
- Inspired by OpenAI's 10,000-agent run, dev open-sources a crowdsourced agent problem-solving platform — Benjaminsen · 2026-09-11
- Lucid: open-source Mac app keeps your laptop awake only while AI agents run — Pitiful_Hedgehog_600 · 2026-09-11
- banteg's snail project crowdsources AI agents to finish matching Snail Mail's 20 remaining functions — banteg · 2026-09-11
- Alex Townsend posts 200 open problems in numerical linear algebra for humans and AI agents — IgorCarron · 2026-09-11
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11