DataRobot for Developers: Skills, MCP, and the agentic developer surface

2026-05-19 · Source: Blog | DataRobot · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, medium

Summary

DataRobot has launched a new developer surface designed to streamline the creation, deployment, and monitoring of production-grade AI agents directly within integrated development environments. This platform integrates four core components: DataRobot Skills, the Global MCP, agent templates, and the LLM Gateway. DataRobot Skills allows developers to inject DataRobot's capabilities, such as AutoML and monitoring, into coding agents like Cursor and VS Code Copilot using natural language commands. The Global MCP, auto-deployed to every DataRobot instance, enables agents to dynamically discover and utilize tools, eliminating the need to embed tool code within the agent itself. Furthermore, datarobot-agent-templates provide scaffolds for frameworks like LangGraph, complete with Pulumi infrastructure and OpenTelemetry tracing for governed deployments. The LLM Gateway offers an OpenAI-compatible endpoint for model access, centralizing governance and credentialing. This integrated approach allows a platform engineer to deploy a governed LangGraph agent with monitoring and tracing rapidly.

Key takeaway

For AI Engineers building production agents, DataRobot's integrated platform significantly reduces plumbing overhead. You can utilize its Skills for direct IDE access to DataRobot capabilities and employ the Global MCP to dynamically manage agent tools. This allows you to focus on agent logic, not infrastructure. Use Agent Assist to design and validate agent behavior before deployment, ensuring governed, scalable, and observable agents are shipped faster.

Key insights

DataRobot simplifies AI agent development by integrating core services directly into developer workflows.

Principles

• Decouple agent logic from tool code.
• Centralize LLM access for governance.
• Validate agent behavior pre-deployment.

Method

Use dr assist to specify agents in natural language, simulate tool-calling for validation, then scaffold with templates for governed deployment via dr task run deploy.

In practice

• Install datarobot-agent-skills for IDE-native DataRobot access.
• Configure Global MCP for dynamic tool discovery.
• Use datarobot-agent-templates for CrewAI/LangGraph scaffolds.

Topics

• AI Agents
• DataRobot Platform
• LLM Gateway
• MLOps Tools
• Agent Assist
• LangChain Integration

Code references

• datarobot-oss/datarobot-agent-skills
• datarobot-community/af-component-datarobot-mcp
• datarobot-community/datarobot-agent-templates

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

Related on AIssential

• Build with Cursor and deploy production-ready AI agents on DataRobot — Modular context packages (Skills) bridge the gap between AI agent prototypes and production deployments on specialized platforms.
• How to design and run an agent in rehearsal – before building it — DataRobot Agent Assist accelerates AI agent development from design to deployment via a natural language CLI.
• How to correctly use MCP servers with your AI Agents — Careful MCP server integration prevents context bloat and optimizes AI agent performance and cost.
• Building Production-Ready AI Agents with Agent Development Kit — ADK simplifies building scalable, production-ready multi-agent AI systems by abstracting orchestration and integrating cloud services.
• Agent Skills — Part 2: Bridging Skills with Production Tool Ecosystems — SkillRegistry decouples tool discovery from execution, enabling scalable agent skill management with existing production tools.
• AI Agents, Tools, MCP, and Skills: The Core, The Embellishment, and The Gimmick — AI agent development requires distinguishing core components like Agents and Tools from less practical elements like MCP.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Blog | DataRobot.