Meta launches Muse personal agent with per-instance VMs and a Sentinel action checker
shengjia_zhao · x · 2026-09-09
Meta launched Muse, its always-on personal AI agent that launches subagent swarms, builds its own tools, and edits itself. Security architecture took most of the effort: each Muse runs in its own secure VM, a separate Sentinel the agent cannot override checks every outbound action, it never sees real credentials, and asks before anything sensitive. The model was trained for zero-shot CLI/skills tool calling, long context, prompt-injection-aware long-trajectory instruction following, and multi-agent coordination, then hardened via dogfooding, agentic red teaming, and a private bug bounty program.
Related event: Meta Launches Muse, an Always-On Personal AI Agent(68 posts)→
More from coding & agent
- DeepMind paper: LLMs can't sample randomly on their own, only convert given seeds — rohanpaul_ai · 2026-10-27
- DeepMind paper: LLMs can't sample randomly on their own — pass random numbers from code instead — rohanpaul_ai · 2026-10-27
- BELL: Open-Source Stack Combines Browserbase, Eve, iMessage and Stripe Link for Personal Agents — jeff_weinstein · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Two engineers rebuilt a FedEx supplier's billing core in 3.5 months vs an estimated 7-8 — alex_verem · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27