Metaflow MCP
Manage ML workflows with AI via MCP. Build Metaflow DAGs, monitor runs, manage artifacts, and deploy data science pipelines.
What is Metaflow MCP?
Metaflow MCP connects AI to Netflix's Metaflow framework. Metaflow makes it easy to build and manage real-life data science projects, with native support for AWS infrastructure and a human-centric API.
Data Science Workflow Intelligence
AI models can monitor running flows, query step artifacts, analyze execution patterns, and help debug failing steps — making data science workflow management conversational.
Configuration
{
"mcpServers": {
"metaflow": {
"command": "npx",
"args": ["metaflow-mcp"],
"env": {
"METAFLOW_SERVICE_URL": "http://localhost:8080"
}
}
}
}
Use Cases
Metaflow MCP serves data science teams building production ML pipelines, organizations using AWS for ML infrastructure, and teams needing AI-powered workflow monitoring and debugging.
Key Features
- Monitor Metaflow run status and logs
- Query artifacts and data from steps
- Manage flow versions and tags
- Track resource utilization per step
- Deploy flows to AWS Step Functions
- Analyze flow execution patterns
Similar MCP Servers
View all →Everything Claude Code
The agent harness performance optimization system.
Mcp For Beginners
This open-source curriculum introduces the fundamentals of MCP.
DesktopCommanderMCP
MCP server for Claude with terminal control and file search.
Docker Hub MCP
Official MCP server to interact with Docker Hub.