XState v6 Beta Launches: State Machines Rewritten for the AI Agent Era
mattpocockuk · x · 2026-11-17
- Stately released the first beta of XState v6 (npm i xstate@beta), a major rewrite focused on making state machines and statecharts easier to read, write, verify, and run anywhere. Core API is considered stable; durable execution remains experimental.
- The project counts 30k+ GitHub stars and 8M+ weekly downloads, used in production for UI logic, backend workflows, and increasingly AI agent orchestration. XState is MIT-licensed, zero-dependency, and built on state machines, statecharts, and the actor model.
- The release is explicitly shaped by the AI era: agents turn out to be excellent at writing XState machines, thanks to years of docs and examples and concepts unchanged since Harel's 1987 statecharts paper.
- mattpocockuk also shared an agent evaluation prompt: have an agent research the beta, find the gnarliest problem it could solve in your app, build a proof-of-concept in a draft PR, verify results, and grade the API on nuance (fewer choices = better agent performance).
More from coding & agent
- Pageloop Launches a Docs Agent That Auto-Captures Screenshots and Updates Stale Docs — prasanna_says · 2026-11-17
- OutfitterBench tests LLM agents on 200-day rental business, exposing commitment failures — Jie Huang · 2026-11-17
- Craftax framework evolves code policies via feedback cascade, reaching final stage 50% of time — Víctor Gallego · 2026-11-17
- Entity disambiguation: sentence context hits 97.9% vs 63.5% for name matching — YvesMulkers · 2026-11-17
- Matt Pocock shares a prompt for having AI agents evaluate new tech via draft PRs — mattpocockuk · 2026-11-17
- Motion launches Claude Code for Video Editing: raw footage plus one prompt, auto-edited — CurieuxExplorer · 2026-11-17