Google Paper: Replace Agent Chat History With Explicit State, Cut Tokens 16x
rohanpaul_ai · x · 2026-08-29
A new Google paper, SKILL.state, argues long-horizon agents need far less conversation history than we give them: swapping the ever-growing transcript for a small explicit execution state cuts tokens and often improves accuracy.
At each step the model receives only skill instructions, a structured state, and the latest observation; reasoning traces are discarded after producing a validated state update, keeping prompt size roughly constant. On a 100-step warehouse task, Gemini-3-Flash with SKILL.state scored 0.94 using 65,408 tokens vs 0.91 and 1,062,387 tokens for a LangGraph-style baseline — a 16.2x difference. Boundary: everything needed later must fit in the state schema.
Related event: Google's SKILL.state Cuts Agent Token Use by 94% with Explicit State(3 posts)→
More from coding & agent
- New paper: agents must "forget" to keep improving in days-long autoresearch runs — matei_zaharia · 2026-10-16
- Turning code into knowledge animations: 4 prompt styles plus a take on Skills vs prompts — vista8 · 2026-10-16
- Neon Realtime turns any Postgres query into a live subscription, built inside the database — matei_zaharia · 2026-10-16
- Tool lets AI agents use your accounts without ever seeing your login credentials — cneuralnetwork · 2026-10-16
- Claude Code plugin /ambition rewrites your task into the version a more ambitious person would ask for — jakedahn · 2026-10-16
- arXiv paper "Self-Evolving Agent Harness" formalizes harness self-evolution with admission rules — rohanpaul_ai · 2026-10-16