aiqbench analysis
TechPulse take: This research introduces JAZ, a minimalist agent framework that challenges the necessity of complex external systems like dedicated memory or file storage by leveraging a single recursive LLM primitive. The study demonstrates that this streamlined approach can outperform specialized frameworks on long-horizon tasks and self-improvement benchmarks while significantly reducing computational costs. It represents a significant step toward simplifying agent architecture without sacrificing capability, focusing purely on the core mechanics of language-model agency.
Harness as a Language: A Minimalist Agent Framework With Maximal Expressivity
arXiv:2609.26891v1 Announce Type: new Abstract: Modern language-model agents are built around the \textit{agent loop}, where the LLM is placed in an environment exposing a set of tools, and the LLM has full control over the workflow by alternating between tool calls and observing their output. However, certain workflows currently require additional engineering beyond the agent loop itself, such as memory systems and self-improving systems. We built an LLM agent framework, JAZ, to explore the extent to which a minimal harness that is little more than the agent loop itself can accomplish tasks these specialized systems are built for. JAZ exposes a single LLM-based primitive invoke and provides a set of built-in hooks that allow the programmer to apply constraints and monitoring. Generalizing existing code-mode agent loops, \texttt{invoke} is the simplest loop that satisfies two defining properties: (1) the LLM can write arbitrary executable code that can include recursive \texttt{invoke}; (2) everything visible to the LLM --- all inputs to \texttt{invoke} as well as its interaction history with the code environment --- are variables in the code environment. We motivate our design from first principles, viewing \texttt{invoke} as a language primitive representing a function whose implementation is provided at runtime by an LLM every time it is called. To validate the design of our core \texttt{invoke} primitive, we evaluate \texttt{invoke} --- with only prompting, no manually designed tools, harness, or external systems (e.g., memory or the file system) --- on workflows traditionally implemented through specialized external harnesses. On long-horizon workflows requiring recall beyond the context window, JAZ invoke outperforms Letta (MemGPT) by 8\% at half its cost on the recall-heavy portion of StuLife. On continual self-improvement, JAZ invoke outperforms ACE by 4\% at a lower cost on AppWorld.
Abstract:Modern language-model agents are built around the \textit{agent loop}, where the LLM is placed in an environment exposing a set of tools, and the LLM has full control over the workflow by alternating between tool calls and observing their output. However, certain workflows currently require additional engineering beyond the agent loop itself, such as memory systems and self-improving systems. We built an LLM agent framework, JAZ, to explore the extent to which a minimal harness that is little more than the agent loop itself can accomplish tasks these specialized systems are built for. JAZ exposes a single LLM-based primitive invoke and provides a set of built-in hooks that allow the programmer to apply constraints and monitoring. Generalizing existing code-mode agent loops, \texttt{invoke} is the simplest loop that satisfies two defining properties: (1) the LLM can write arbitrary executable code that can include recursive \texttt{invoke}; (2) everything visible to the LLM --- all inputs to \texttt{invoke} as well as its interaction history with the code environment --- are variables in the code environment. We motivate our design from first principles, viewing \texttt{invoke} as a language primitive representing a function whose implementation is provided at runtime by an LLM every time it is called. To validate the design of our core \texttt{invoke} primitive, we evaluate \texttt{invoke} --- with only prompting, no manually designed tools, harness, or external systems (e.g., memory or the file system) --- on workflows traditionally implemented through specialized external harnesses. On long-horizon workflows requiring recall beyond the context window, JAZ invoke outperforms Letta (MemGPT) by 8\% at half its cost on the recall-heavy portion of StuLife. On continual self-improvement, JAZ invoke outperforms ACE by 4\% at a lower cost on AppWorld.
Artificial Intelligence (cs.AI)
arXiv:2609.26891 [cs.AI]
(or arXiv:2609.26891v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.26891 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 · 376 words
Related coverage
Links #5: 2026/07
LessWrong · Jul 31, 2026
Capturing token IDs during agentic interactions for better reinforcement learning
Amazon Science · Jul 9, 2026
Humanity’s Last Invention — Richard Socher of Recursive
Latent Space · Sep 14, 2026
Deconfusing Corrigibility
AI Alignment Forum · Oct 17, 2026