Request: ReAct and Plan & Execute as Dedicated Nodes (Similar to Text Classifier Node)

The concept is:

Re-integrate the ReAct Agent and Plan and Execute Agent as dedicated nodes, similar to the current Information Extractor or Text Classifier nodes. These nodes would cater to specific use cases such as multi-step reasoning, planning, and dynamic tool utilization, simplifying their implementation without requiring manual LangChain code.

My use case:

I have been developing a self-assisting AI agent utilizing cascaded agents like Plan and Execute and ReAct. These agents were instrumental in creating autonomous behaviors, enabling the AI to break down tasks, reason through steps, and dynamically invoke tools to achieve complex objectives. Following the deprecation of the “Agent Type” field, replicating this functionality now demands significantly more effort using lower-level components.

I believe adding this would be beneficial because:

• Reintroducing these as dedicated nodes would establish a good balance—maintaining simplicity for the majority of users, while still empowering advanced users to construct intelligent agents with minimal effort.

Are there any resources to support this?

• Deprecation of Agent Type in AI Agent Node – GitHub PR #13459
• Examples demonstrating how Information Extractor and Text Classifier nodes make advanced tasks user-friendly.

Are you willing to contribute to this?

I am happy to provide more details about my use case, test new features, or contribute ideas to help shape the development of these nodes.