Importing LangGraph Agents in watsonx Orchestrate

2026-03-08 · Source: Niklas Heidloff · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

IBM's watsonx Orchestrate now features an experimental capability to import LangGraph agents, enhancing its multi-agent system platform for enterprise integrations. These imported agents run in isolated environments within Orchestrate and can function as top-level or collaborator agents, offering flexibility in defining system prompts and instructions. While direct access to other Orchestrate artifacts like tools and knowledge bases is currently limited, and agents must not be compiled, this integration supports simple messages and Python. LangGraph agents can also be integrated via A2A for long-running operations and asynchronous callbacks. The process involves using the Orchestrate ADK (Agent Developer Kit) CLI to create custom agents from `agent.yaml`, `agent.py`, and `requirements.txt` files.

Key takeaway

For AI Engineers building collaborative systems on IBM watsonx Orchestrate, you can now integrate custom LangGraph agents directly. This allows for greater control over agent behavior and system prompts, though you should be aware of current limitations regarding access to other Orchestrate artifacts. Consider using A2A integration for agents requiring long-running or asynchronous operations.

Key insights

IBM watsonx Orchestrate now supports importing LangGraph agents, enabling flexible, collaborative AI systems.

Principles

• Multi-agent systems drive business value.
• Deep integration with enterprise systems is key.

Method

Import LangGraph agents into watsonx Orchestrate using the ADK CLI, providing `agent.yaml`, `agent.py`, and `requirements.txt` files for deployment.

In practice

• Define custom system prompts for imported agents.
• Use A2A for long-running agent operations.

Topics

• Multi-agent systems
• watsonx Orchestrate
• LangGraph
• Agent integration
• Enterprise AI platforms

Best for: AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Testing Conversations in Orchestrate via Agentic Skill — The `watsonx-orchestrate` skill automates agent testing in live Orchestrate environments, supporting TDD for conversational AI.
• IBM Consulting Expands AI Capabilities to Accelerate Enterprise Transformation — IBM is expanding its AI consulting and platform services to enable controlled, context-aware AI deployment in enterprises.
• Build Cross-Language Multi-Agent Team with Google’s Agent Development Kit and A2A — Cross-language multi-agent systems using ADK and A2A enhance resilience and modularity for production AI.
• IBM Introduces Industry Solutions for AI Powered Experience Orchestration with Adobe — Orchestrating data, decisions, and delivery with AI enables real-time customer responsiveness, significantly improving business outcomes.
• Accessing watsonx Orchestrate from Bob via CLI — Using agentic skills with CLIs offers advantages over MCP tools for AI development agents like IBM Bob.
• IBM and ElevenLabs: Enterprise AI Finds its Voice — The IBM-ElevenLabs partnership enables secure, scalable, voice-first agentic AI for enterprises, transforming human-computer interaction.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Niklas Heidloff.