LangChain launches RubricMiddleware for Deep Agents to let agents self-evaluate and correct against rubrics
Original title:Introducing Rubrics: Build Agents that Evaluate and Correct Their Work
AI summary
LangChain added RubricMiddleware (beta) for Deep Agents: developers pass a rubric checklist at invoke time, and a separate grader sub-agent reviews each item before the agent finishes, optionally calling tools to gather evidence and returning item-by-item feedback. If the rubric is not met, the feedback is injected into the conversation and the agent reruns until the rubric is satisfied or max_iterations is reached.
Key Takeaways
- LangChain added RubricMiddleware (beta) for Deep Agents: developers pass a rubric checklist at…
- optionally calling tools to gather evidence and returning item-by-item feedback.
- If the rubric is not met,the feedback is injected into the conversation and the agent reruns until…
Quick Facts
- Published
- Updated
- Source
- LangChain:Blog(RSS)
- Category
- Products & Apps
- AI score
- 63
原文给出 grader 子智能体、循环终止条件和配置示例,读者可以照此在 Deep Agents 中接入自评纠错流程。
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: LangChain launches RubricMiddleware for Deep Agents to let agents self-evaluate and correct against rubrics. June 3, 2026. https://new.nat6.net/en/news/cms3x0dh503ayro8261lka0b9/