aiqbench analysis
This paper proposes Skillware, a framework that replaces brittle prompt engineering with modular, deterministic software bundles to improve the reliability and efficiency of autonomous agents. By decoupling task definitions from execution logic, the approach significantly reduces token usage and inference costs while enabling smaller local models to outperform larger frontier models in complex workflows. This shift represents a critical evolution in AI architecture, moving away from context-heavy prompting toward structured, auditable capability execution essential for production-grade logical systems.
Don't Prompt Your Agents. Equip Them: Skillware as a Modular Standard for Sovereign, Auditable, and Deterministic Capability Execution for Logical Systems
arXiv:2611.07652v1 Announce Type: new Abstract: Current autonomous logical systems rely on a fragile assumption: that language models should dynamically write code, guess API payloads, and memorize operational constraints through prompt engineering. In practice, this burns massive context windows, inflates API bills, and produces brittle agents that fail unpredictably in production. In this work, we present Skillware, an open architectural framework and capability registry that formally separates what an agent wants to do from how that task is executed. Instead of prompting models with loose text instructions or markdown files baptized as 'skills', Skillware equips them with modular, versioned software capability bundles. Much like skill hotkeys in video games that execute deterministically under fixed world rules, each bundle combines a typed JSON input contract, deterministic Python code, task directives, bundled datasets, and automated unit tests. In-process adapters translate these capabilities into native tool declarations for any model, from frontier cloud APIs to local open-weight runtimes with zero network latency. Evaluating 23 frontier and open-weight models across 368 benchmark conditions, we show that small local models (3B to 8B parameters) equipped with Skillware consistently outperform unequipped frontier models on complex multi-step workflows. They cut token usage by up to 84.8%, slash inference costs by up to 97.3%, and eliminate parameter errors entirely. True agent reliability does not come from larger models or longer prompts, but from equipping logical systems with verified, deterministic software. Finally, we articulate the trajectory of capability bundles across an invariant continuum of information processing systems: from digital logical systems today, to IoT, drones, and embodied robotics tomorrow, establishing a foundation for auditable, real-time capability execution.
Title:Don't Prompt Your Agents. Equip Them: Skillware as a Modular Standard for Sovereign, Auditable, and Deterministic Capability Execution for Logical Systems
Abstract:Current autonomous logical systems rely on a fragile assumption: that language models should dynamically write code, guess API payloads, and memorize operational constraints through prompt engineering. In practice, this burns massive context windows, inflates API bills, and produces brittle agents that fail unpredictably in production. In this work, we present Skillware, an open architectural framework and capability registry that formally separates what an agent wants to do from how that task is executed. Instead of prompting models with loose text instructions or markdown files baptized as 'skills', Skillware equips them with modular, versioned software capability bundles. Much like skill hotkeys in video games that execute deterministically under fixed world rules, each bundle combines a typed JSON input contract, deterministic Python code, task directives, bundled datasets, and automated unit tests. In-process adapters translate these capabilities into native tool declarations for any model, from frontier cloud APIs to local open-weight runtimes with zero network latency. Evaluating 23 frontier and open-weight models across 368 benchmark conditions, we show that small local models (3B to 8B parameters) equipped with Skillware consistently outperform unequipped frontier models on complex multi-step workflows. They cut token usage by up to 84.8%, slash inference costs by up to 97.3%, and eliminate parameter errors entirely. True agent reliability does not come from larger models or longer prompts, but from equipping logical systems with verified, deterministic software. Finally, we articulate the trajectory of capability bundles across an invariant continuum of information processing systems: from digital logical systems today, to IoT, drones, and embodied robotics tomorrow, establishing a foundation for auditable, real-time capability execution.
51 pages, 26 figures. Code repository and benchmarks available at this https URL
Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
I.2.11; D.2.11; I.2.7
arXiv:2611.07652 [cs.AI]
(or arXiv:2611.07652v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2611.07652 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 396 words
Related coverage
The Three-Ring Architecture: Governing Agents in the Era of On-Platform Organisations
arXiv · Jun 8, 2026
Stochastic CHAOS: Why Deterministic Inference Kills, and Distributional Variability Is the Heartbeat of Artifical Cognition
arXiv · Oct 15, 2026
ToolFence: Fine-Grained Authorization for Secure Tool-Using LLM Agents
arXiv · Sep 30, 2026
ReliCAD: From Uncertain LLM Generation to Reliable Parametric CAD Modeling
arXiv · Sep 22, 2026