LangChain adds human-in-the-loop pauses to Managed Deep Agents in public beta

LangChain has launched human-in-the-loop support in public beta for Managed Deep Agents, according to a LangChain post on X. An agent can pause to ask the user a question, and the question appears with buttons in the Slack thread.

The agent resumes once the user clicks one of the buttons.

The only available report is LangChain's own announcement, which describes the feature as in public beta. No pricing, availability limits, or independent testing has been reported.