Two methodology contributions are presented: a declarative-markdown harness with a small mutation surface, and an auto-research loop in which failure-mode analysis on completed experiments proposes new hypotheses, optionally human-reviewed, that progress through the same experiment workflow.
Comparisons against stronger model and coding-agent competitors further indicate that both domain-specific agent runtime structure and foundation-model strength matter for autonomous data analysis.
AgentRadio is presented, an asynchronous message-passing layer that equips coding-agent harnesses with three primitives: threads, messages, and waiting for mentions that shows the gain growing with task difficulty, consistent with mid-course correction as the underlying mechanism.
Xinxing Ren, Qianbo Zang, Ziyan Wang et al.· arXiv.org· 0 citations
This work presents MetaRoute-Bench, an open, inspectable framework for comparing meta-decision policies under a shared execution model, and releases task generation, policies, traces, tests, and analysis artifacts to support live-system validation.
Natan Vidra, Alina Kapanova, Arun Kanhai et al.· 0 citations
It is demonstrated that communication format and implementation details determine outcomes more than architectural complexity, and that simpler approaches match or outperform multi-agent pipelines for local 7B model deployment.
Ashish Prajapati, Om Mohite· arXiv.org· 0 citations
Testing three production engines showed that one achieves 22/22 syntactic TPC-H acceptance but falls short on full semantic validation, function-level pass rates of 94% on non-string categories mask dramatic per- category swings.
This survey treats isolation as a first-class principle for LLM-agent system safety, and organizes the literature with a boundary-centric taxonomy of five boundaries: user-agent, agent-tool, agent-execution, agent-agent, and system-environment.
