130-Page Survey: State and Lifecycle of Always-On Agents
omarsar0 · x · 2026-07-06
A new 130+ page survey systematically reviews "always-on agents"—systems whose future behavior relies on persistent state accumulated across multiple interactions. It treats "state" as a concept far broader than just "memory." The review incorporates task ledgers, permissions, credentials, commitments, provenance, and triggers into the state category. It scores each state type across six dimensions: authority, scope, mutability, provenance, recoverability, and actionability, covering the entire lifecycle from writing, retrieval, and forgetting to auditing and revocation.
More from coding & agent
- Agent verified 'success' after double-voting: a lesson in harness vs model bugs — TejasKumar_ · 2026-10-26
- diginomica: agentic automation hype is overdone, and 'The Hackening' of enterprise security has begun — jonerp · 2026-10-26
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Devin launches Security suite powered by Claude Mythos with $100M+ credits — DevinAI · 2026-10-26
- HackBench injects reward hacks into agent trajectories to stress-test monitors — DanielKhashabi · 2026-10-26
- Hugging Face ships hf CLI updates: deduped cache saves 19GB, faster buckets, retryable GPU jobs — yb2698 · 2026-10-26