us-all/mlflow-mcp-server
🔌 MCP Serverus-all
Connect your AI agents to MLflow for seamless experiment tracking, model management, and GenAI trace analysis.
The mlflow-mcp-server is a comprehensive integration layer designed to bring MLflow's robust MLOps capabilities directly into the context of AI agents. By implementing the Model Context Protocol, it allows LLM-based agents to programmatically query and manipulate MLflow entities without leaving their native environment. Key features include full CRUD support for experiments and runs, deep integration with registered models, and advanced support for MLflow 3 GenAI traces. The server includes five built-in workflow prompts: 'analyze-failed-traces', 'promote-best-run', 'compare-top-runs', and others, which automate complex MLOps tasks. This tool is essential for teams looking to build autonomous agents that can monitor training pipelines, evaluate model performance, and manage deployment lifecycles through natural language commands or agentic reasoning.
💡Highlights
- ├─82 tools for MLflow management
- ├─MLflow 3 GenAI trace support
- └─Built-in automated workflow prompts
🎯For
- ├─MLOps Engineers
- └─AI Agents Developers