Mcp Zenml
Connects Claude to your ZenML server's REST API to browse and manage ML pipelines. You get read access to the full ZenML object model: stacks, components, pipeline runs, artifacts, deployments, model versions, and more. The real workflow enabler is triggering new pipeline runs directly from chat using snapshots (the current approach) or the older run templates. It also surfaces deployment logs and step-level details, so you can debug failed runs or check serving endpoint status without leaving your IDE. Best for teams already running ZenML who want LLM-assisted pipeline operations and infrastructure introspection. Ships as a Docker container, connects via your ZenML server URL and API key.
- GitHub stars
- 49
- tools
- 16
- auth
- Secret needed
- transport
- STDIO
- env vars
- 4
- last updated
- Nov 6, 2026
Tools
16 toolsPublic tool metadata for what this MCP can expose to an agent.
zenml_describe_resourcesDiscover supported resource types and bounded operation schemas
Discover supported resource types and bounded operation schemas
No parameter schema in public metadata yet.
zenml_list_resourcesList one resource type with validated filters and pagination
List one resource type with validated filters and pagination
No parameter schema in public metadata yet.
zenml_get_resourceGet one resource, with parent and project scope where required
Get one resource, with parent and project scope where required
No parameter schema in public metadata yet.
zenml_create_resourceCreate a supported resource from a typed payload
Create a supported resource from a typed payload
No parameter schema in public metadata yet.
zenml_update_resourceUpdate one exact resource UUID
Update one exact resource UUID
No parameter schema in public metadata yet.
zenml_delete_resourceDelete or archive one exact resource UUID
Delete or archive one exact resource UUID
No parameter schema in public metadata yet.
zenml_action_resourceRun an allowlisted lifecycle or relation action without retries
Run an allowlisted lifecycle or relation action without retries
No parameter schema in public metadata yet.
diagnose_zenml_setupDiagnostics for the ZenML setup
Diagnostics for the ZenML setup
No parameter schema in public metadata yet.
get_active_userActive user context
Active user context
No parameter schema in public metadata yet.
get_active_projectActive project context
Active project context
No parameter schema in public metadata yet.
trigger_pipelineRun pipeline execution
Run pipeline execution
No parameter schema in public metadata yet.
get_step_logsStreamed step logs, returning at most 50,000 entries with a possibly_truncated flag
Streamed step logs, returning at most 50,000 entries with a possibly_truncated flag
No parameter schema in public metadata yet.
get_step_codeStreamed step code
Streamed step code
No parameter schema in public metadata yet.
get_deployment_logsStreamed deployment logs
Streamed deployment logs
No parameter schema in public metadata yet.
open_pipeline_run_dashboardInteractive table of recent pipeline runs with status, step details, and logs
Interactive table of recent pipeline runs with status, step details, and logs
No parameter schema in public metadata yet.
open_run_activity_chartBar chart of pipeline run activity over the last 30 days with status breakdown
Bar chart of pipeline run activity over the last 30 days with status breakdown
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
ZENML_STORE_URL*Base URL of your ZenML server (e.g., https://<workspace-id>-zenml.cloudinfra.zenml.io).
ZENML_STORE_API_KEY*secretAPI key used to authenticate with your ZenML server (ideally a service account key).
ZENML_MCP_PROFILETool registration profile: compact (default) or legacy.
ZENML_MCP_WRITE_POLICYMutation policy: read_write (default) or read_only.
Readme
MCP Server for ZenML
This project implements a Model Context Protocol (MCP) server for interacting with the ZenML API.
ZenML MCP Server
What is MCP?
The Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to Large Language Models (LLMs). It acts like a "USB-C port for AI applications" - providing a standardized way to connect AI models to different data sources and tools.
Details
- Repository
- github.com/zenml-io/mcp-zenml
- Package
docker.io/zenmldocker/mcp-zenml:2.1.1- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active