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 authors introduce a new bounded contract for long-horizon Large Language Model (LLM) agents in the game Slay the Spire 2. The contract allows decisions to be made from a fresh user message, with no raw cross-decision transcript appended. The authors demonstrate that this contract improves performance when triggered strategic skills are enabled, but further investigation is needed for a statistically decisive comparison.

What the Document Says

• The text discusses a new contract for long-horizon LLM agents in the game Slay the Spire 2.
• The contract allows decisions to be made from a fresh user message, with no raw cross-decision transcript appended.
• The authors report that a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled.
• The authors release a reproducible testbed with 298 completed trajectories, frozen memory/skill snapshots, prompt records, and analysis scripts.

Key Concepts Mentioned

• Long-horizon LLM agents
• Bounded contract
• Slay the Spire 2
• Game decisions
• Strategic skills

Claims Made by the Document

• The new contract improves performance when triggered strategic skills are enabled.

Evidence Used

• Fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled.

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.