SKILL.state: Scalable Long-Horizon Agent Skills

• URL: http://arxiv.org/abs/2608.26263v3
• Date: Wed, 02 Sep 2026 02:32:58 GMT
• Title: SKILL.state: Scalable Long-Horizon Agent Skills
• Abstract summary: We present SKILL, a runtime architecture that replaces append-only conversational history with an explicit, mutable execution state.<n>Our results demonstrate that explicit execution state is an effective and architecture-agnostic abstraction for scalable long-horizon agent skills.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) increasingly act as autonomous agents executing complex, long-running procedural skills. Existing agent runtimes maintain execution by continually appending observations, actions, and intermediate reasoning traces to an ever-growing conversation history, causing latency degradation and context-poisoning failures over long horizons. We present SKILL. state, a runtime architecture that replaces append-only conversational history with an explicit, mutable execution state. At each execution step, the model receives only the immutable skill specification, the current structured execution state, and the latest observation. Intermediate reasoning is discarded immediately after producing a validated state update, preventing prompt growth with execution history. Across diverse datasets, models, and execution environments, SKILL. state improves task accuracy while substantially reducing cumulative token consumption. Our results demonstrate that explicit execution state is an effective and architecture-agnostic abstraction for scalable long-horizon agent skills.

Related papers

• Turning Interaction History into Execution State: A Runtime Layer for Long-Horizon Coding Agents [48.967927222079965]
  Ledger is a runtime layer that distills an agent's completed interactions into an explicit execution state.<n> Ledger keeps this state in an online execution ledger and applies it at two boundaries of every step.<n>Across all 500 SWE-bench Verified instances, Ledger raises Pass@1 from 56.2% to 64.2% with GPT-5 mini and from 75.8% to 81.0% with MiniMax M2.5.
  arXiv  Detail & Related papers  (2026-08-01T18:25:00Z)
• PhyAgentOS: A Self-Evolving Operating System for Embodied Agents with Decoupled Cognitive Planning and Physical Execution [49.776611937968]
  We present PhyAgentOS, a foundation delivering scheduling, verification, memory, benchmarking, and safety as system-level services.<n>Its Session-Centered treats a session, not an action, as the minimum unit of scheduling, compatibility, supervised execution, evidence collection, and acceptance.<n>SessionVerifier distinguishes execution termination from semantic task completion via evidence-grounded verdicts of success, failure, or replan.<n> Benchmarking reuses the deployment session and verification path, so results trace to real execution.
  arXiv  Detail & Related papers  (2026-07-18T04:46:53Z)
• StructAgent: Harness Long-horizon Digital Agents with Unified Causal Structure [29.25570731801935]
  StructAgent is a state-centered framework for maintaining compact, verifiable task progress.<n>It enables explicit progress checkpointing, evidence-driven task completion, targeted failure recovery, and tool-supported execution.<n>It generalizes beyond desktop environments to Minecraft, demonstrating the generality of our design.
  arXiv  Detail & Related papers  (2026-07-13T10:48:25Z)
• ContextFlow: Hierarchical Task-State Alignment for Long-Horizon Embodied Agents [41.517468049737936]
  Long-horizon embodied agents increasingly delegate navigation, search, approach, and manipulation to specialist executors.<n>We study task-state misalignment, a task-level consistency failure in which the planner's active stage, runtime evidence, remembered context, and delegated executor no longer justify the same next-step decision.<n>We propose ContextFlow, an inspectable alignment framework that represents stages as explicit contracts, converts runtime observations into evidence packets, and applies scoped updates.
  arXiv  Detail & Related papers  (2026-05-19T03:49:17Z)
• ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents [59.626170560327274]
  We present textbfClawForge, a generator-backed benchmark framework for executable command-line categories under state conflict.<n>We instantiate this framework as the ClawForge-Bench (17 scenarios, 6 ability categories)
  arXiv  Detail & Related papers  (2026-05-13T21:34:08Z)
• AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents [50.481033105867205]
  We introduce AgentProcessBench, the first benchmark dedicated to evaluating step-level effectiveness in realistic, tool-augmented trajectories.<n>The benchmark comprises 1,000 diverse trajectories and 8,509 human-labeled step annotations with 89.1% inter-annotator agreement.<n>It features a ternary labeling scheme to capture exploration and an error propagation rule to reduce labeling ambiguity.
  arXiv  Detail & Related papers  (2026-03-15T16:13:58Z)
• Agents Learn Their Runtime: Interpreter Persistence as Training-Time Semantics [4.774525456207306]
  Tool-augmented LLMs are increasingly deployed as agents that interleave natural-language reasoning with executable Python actions.<n>We ask whether interpreter persistence is merely a runtime scaffold, or a property of the training data that shapes how agents learn to use the interpreter.<n>We introduce Opaque Knapsack, a procedurally generated family of partially observable optimization tasks designed to prevent one-shot solutions.
  arXiv  Detail & Related papers  (2026-03-01T18:08:02Z)
• Towards Structured, State-Aware, and Execution-Grounded Reasoning for Software Engineering Agents [20.481392033796265]
  We argue that we need to move beyond reactive behavior toward a structured, state-aware, and execution-grounded reasoning.<n>We outline how explicit structure, persistent and evolving state, and the integration of execution-grounded feedback can help SE agents perform more coherent and reliable reasoning.
  arXiv  Detail & Related papers  (2026-02-04T15:07:53Z)
• Haste Makes Waste: Evaluating Planning Abilities of LLMs for Efficient and Feasible Multitasking with Time Constraints Between Actions [56.88110850242265]
  We present Recipe2Plan, a novel benchmark framework based on real-world cooking scenarios.<n>Unlike conventional benchmarks, Recipe2Plan challenges agents to optimize cooking time through parallel task execution.
  arXiv  Detail & Related papers  (2025-03-04T03:27:02Z)
• CItruS: Chunked Instruction-aware State Eviction for Long Sequence Modeling [52.404072802235234]
  We introduce Chunked Instruction-aware State Eviction (CItruS), a novel modeling technique that integrates the attention preferences useful for a downstream task into the eviction process of hidden states. Our training-free method exhibits superior performance on long sequence comprehension and retrieval tasks over several strong baselines under the same memory budget.
  arXiv  Detail & Related papers  (2024-06-17T18:34:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.