AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Xiangchen Cheng, Yunwei Jiang, Jianwen Sun, Zizhen Li, Chuanhao Li, Xiangcheng Cao, Yihao Liu, Fanrui Zhang, Li Jin, Kaipeng Zhang
• Venue: arXiv: cs.AI
• Published: 2026-07-02
• Collected: 2026-07-03
• URL: http://arxiv.org/abs/2607.02255v1
• PDF: https://arxiv.org/pdf/2607.02255v1
• DOI: UNKNOWN

Plain-Language Brief

The researchers introduce a new contract system for long-term learning machine (LLM) agents that limits the memory of each decision to a fresh user message, assembled by typed retrieval. They test this contract in the game Slay the Spire 2 and compare its performance with a no-store baseline and public accumulating-context baselines. The researchers release a reproducible testbed for their methodology.

What the Document Says

• The paper introduces a new bounded memory contract for LLM agents in which each decision is made from a fresh user message assembled by typed retrieval, with no raw cross-decision transcript appended.
• The contract is tested on Slay the Spire 2, a closed-rule stochastic deck-building game.
• A fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10.
• The task is considered hard but not saturated.

Key Concepts Mentioned

• long-horizon LLM agent
• bounded memory contract
• Slay the Spire 2
• fixed-A0 ablation

Claims Made by the Document

• Adding the skill layer to the no-store baseline increases the win rate in Slay the Spire 2.

Evidence Used

• Within our harness, a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Memory for a long-horizon LLM agent is a contract about what each future decision is allowed to see. The simplest contract appends past observations, tool calls, and reflections to every prompt, which makes prior context easy to access but also turns it into a jumbled mixture in which the effect of any single memory component is hard to isolate. We introduce and instrument an alternative bounded contract: every decision is made from a fresh user message assembled by typed retrieval, with no raw cross-decision transcript appended. The prompt thus stays bounded across runs of any length, and any single layer can be ablated in isolation. We instantiate the contract in Slay the Spire 2, a closed-rule stochastic deck-building game whose runs require hundreds of tactical and strategic decisions. A public online benchmark of frontier LLMs on the same game reports zero wins at the lowest difficulty across five configurations, and the developer-reported human win rate at the same difficulty is 16%; the task is hard but not saturated. Within our harness, a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10. At this sample size the comparison is directional rather than statistically decisive (Fisher exact p\approx0.37); a cross-backbone probe and public accumulating-context baselines are reported as operational comparisons rather than controlled tests of the contract variable itself. We release a reproducible testbed: 298 completed trajectories with condition tags, frozen memory/skill snapshots, prompt records, and analysis scripts -- an agent design and a validated, reusable methodology for studying how explicit memory layers shape long-horizon LLM-agent decisions.