{"slug": "auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied", "title": "Auriga Technical Report: A Task-Agnostic Agent Harness for Zero-Shot Embodied Manipulation", "summary": "Auriga's technical report introduces a task-agnostic agent harness for zero-shot embodied manipulation, addressing the planning-execution gap it measured on RoboDojo, where removing arm execution raises GPT-6 Astra's success rate from 2. The report frames vision-language models as capable of interpreting instructions and planning manipulation tasks but struggling when acting directly as robot policies.", "body_md": "Vision-language models can interpret instructions and plan manipulation tasks, yet struggle when acting directly as robot policies. On RoboDojo, a large planning-execution gap separates high-level task decisions from physical execution: removing arm execution raises GPT-6 Astra's success rate from 2", "url": "https://wpnews.pro/news/auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied", "canonical_source": "https://aiflash.com/news/149249/", "published_at": "2026-11-10 14:30:14+00:00", "updated_at": "2026-11-10 14:53:58.266338+00:00", "lang": "en", "topics": ["robotics", "ai-agents", "ai-research", "large-language-models"], "entities": ["Auriga", "RoboDojo", "GPT-6 Astra"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied", "markdown": "https://wpnews.pro/news/auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied.md", "text": "https://wpnews.pro/news/auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied.txt", "jsonld": "https://wpnews.pro/news/auriga-technical-report-a-task-agnostic-agent-harness-for-zero-shot-embodied.jsonld"}}