Multi-agent LLM workflows -- systems composed of multiple role-specific LLM calls -- often outperform single-prompt baselines, but they remain difficult to debug and refine. Failures can originate from subtle errors in intermediate outputs that propagate to downstream nodes, requiring developers to inspect long traces and infer which agent to modify. We present PROTEA, a unified interface for offline, test-driven improvement of multi-agent workflows. PROTEA executes a workflow, scores intermediate node outputs with configurable rubrics, and overlays per-node states and rationales on the workflow graph to localize likely bottlenecks. To support complex systems where final-answer references are the primary supervision, PROTEA performs backward node evaluation: it generates candidate node-level expectations from final-answer references and graph context, then compares them with observed node outputs. For selected nodes, PROTEA presents targeted prompt revisions as editable before/after comparisons, then automatically reruns and re-evaluates the workflow to show output changes and score trajectories within the same interface. In two production-adjacent workflows, PROTEA improved document-inspection accuracy from 64.3% to 83.9% and recommendation Hit@5 from 0.30 to 0.38. In a formative study with six experienced LLM developers, participants valued graph-level localization, per-node rationales, and editable before/after prompt revisions.

翻译：摘要：多智能体LLM工作流——由多个角色特定LLM调用组成的系统——通常优于单提示基线方法，但其调试与优化仍面临挑战。错误可能源于中间输出的细微偏差，这些偏差会传播至下游节点，迫使开发者检查冗长的执行轨迹并推断需要修改的智能体。我们提出PROTEA，一个用于多智能体工作流离线、测试驱动改进的统一接口。PROTEA执行工作流，通过可配置的评分标准对中间节点输出进行评分，并在工作流图上叠加各节点状态与推理依据，以定位潜在瓶颈。为支持以最终答案参考为主要监督信号的复杂系统，PROTEA执行反向节点评估：从最终答案参考与图上下文生成候选节点级期望，再与观测到的节点输出进行比较。针对选定节点，PROTEA以可编辑的前后对比形式呈现目标提示修订，并自动重新运行与评估工作流，在同一界面中展示输出变化与评分轨迹。在两个生产级工作流中，PROTEA将文档检查准确率从64.3%提升至83.9%，推荐命中率@5从0.30提升至0.38。在针对六位资深LLM开发者的形成性研究中，参与者高度评价了图级定位、节点级推理依据以及可编辑的前后提示修订功能。