DeepJudge Introduces Agent Handoff Protocol for Legal AI Platforms

DeepJudge has introduced Agent Handoff Protocol, a way for users to move work between AI platforms without losing context, according to Legaltech News.

The protocol is designed to carry task context such as source materials, conversation history, prior analysis, objectives, and related resources from one AI product to another. DeepJudge says this is meant for cases where a user needs to continue work inside another product rather than only exchange information through an API or Model Context Protocol connection.

Harvey plans to implement the protocol, with a beta integration set for this month. Thomson Reuters also says it will support the protocol in CoCounsel for Legal, though it has not given timing or implementation details.

DeepJudge is publishing the specification rather than a reference implementation. The protocol defines how two agent applications pass a user, an objective, selected context, and resources between products, with an optional hand back to the original application.