Loading...
The ZenML MCP server provides access to ZenML metadata and pipeline execution, enabling LLMs to interact with and manage ML workflows. Triggering pipelines and accessing secrets pose the highest risks.
Boost this tool
Subscribe to listing upgrades or segmented pushes.
The ZenML MCP server provides access to ZenML metadata and pipeline execution, enabling LLMs to interact with and manage ML workflows. Triggering pipelines and accessing secrets pose the highest risks.
The ZenML MCP server is relatively safe for read-only operations. The greatest risks come from triggering pipelines and accessing secrets. Proper ZenML access controls and monitoring are essential to mitigate these risks.
Performance depends on the ZenML server's responsiveness and network latency. Large queries or log retrievals may take longer.
Cost depends on the underlying ZenML server's infrastructure and resource usage. Triggering pipelines consumes resources and may incur costs.
npx cloudflared tunnel --url http://localhost:8001{
"mcpServers": {
"zenml": {
"command": "/usr/local/bin/uv",
"args": ["run", "path/to/zenml_server.py"],
"env": {
"LOGLEVEL": "INFO",
"NO_COLOR": "1",
"PYTHONUNBUFFERED": "1",
"PYTHONIOENCODING": "UTF-8",
"ZENML_STORE_URL": "https://your-zenml-server-goes-here.com",
"ZENML_STORE_API_KEY": "your-api-key-here"
}
}
}
}get_snapshotRetrieves a specific pipeline snapshot by name or ID.
Read-only access to pipeline configuration metadata.
list_snapshotsLists available pipeline snapshots, optionally filtered by criteria.
Read-only access to pipeline configuration metadata.
get_deploymentRetrieves the runtime status and URL of a deployment.
Read-only access to deployment metadata.
list_deploymentsLists deployments, optionally filtered by status, pipeline, or tag.
Read-only access to deployment metadata.
get_deployment_logsRetrieves logs from a deployment, with configurable tail length.
Read-only access to deployment logs (bounded output).
trigger_pipelineTriggers a new pipeline run using a snapshot or run template.
Initiates pipeline execution, potentially consuming resources.
get_active_projectRetrieves the currently active project.
Read-only access to project metadata.
get_projectRetrieves project details by name or ID.
Read-only access to project metadata.
list_projectsLists all projects.
Read-only access to project metadata.
get_tagRetrieves tag details (exclusive, colors).
Read-only access to tag metadata.
list_tagsLists tags, optionally filtered by resource type.
Read-only access to tag metadata.
get_buildRetrieves build details (image, code embedding).
Read-only access to build metadata.
list_buildsLists builds, optionally filtered by criteria.
Read-only access to build metadata.
get_userRetrieves user details.
Read-only access to user metadata.
list_usersLists all users.
Read-only access to user metadata.
get_active_userRetrieves the currently active user.
Read-only access to user metadata.
get_stackRetrieves stack configuration details.
Read-only access to stack metadata.
list_stacksLists all stack configurations.
Read-only access to stack metadata.
get_stack_componentRetrieves stack component details.
Read-only access to stack component metadata.
list_stack_componentsLists stack components.
Read-only access to stack component metadata.
get_flavorRetrieves component flavor details.
Read-only access to flavor metadata.
list_flavorsLists component flavors.
Read-only access to flavor metadata.
get_service_connectorRetrieves service connector details.
Read-only access to service connector metadata.
list_service_connectorsLists service connectors.
Read-only access to service connector metadata.
get_pipeline_runRetrieves pipeline run details.
Read-only access to pipeline run metadata.
list_pipeline_runsLists pipeline runs.
Read-only access to pipeline run metadata.
get_run_stepRetrieves step details for a pipeline run.
Read-only access to step metadata.
list_run_stepsLists steps for a pipeline run.
Read-only access to step metadata.
get_step_logsRetrieves logs for a specific step in a pipeline run.
Read-only access to step logs.
get_step_codeRetrieves the source code for a specific step in a pipeline run.
Read-only access to step source code.
list_pipelinesLists pipeline definitions.
Read-only access to pipeline definitions.
get_pipeline_detailsRetrieves details for a specific pipeline.
Read-only access to pipeline details.
get_scheduleRetrieves schedule details.
Read-only access to schedule metadata.
list_schedulesLists schedules.
Read-only access to schedule metadata.
list_artifactsLists artifact metadata.
Read-only access to artifact metadata.
list_secretsLists secret names (not values).
Read-only access to secret names.
get_serviceRetrieves model service details.
Read-only access to service metadata.
list_servicesLists model services.
Read-only access to service metadata.
get_modelRetrieves model registry entry details.
Read-only access to model metadata.
list_modelsLists model registry entries.
Read-only access to model metadata.
get_model_versionRetrieves model version details.
Read-only access to model version metadata.
list_model_versionsLists model versions.
Read-only access to model version metadata.
open_pipeline_run_dashboardOpens an interactive pipeline runs dashboard.
Opens a dashboard within a sandboxed iframe.
open_run_activity_chartOpens a 30-day run activity bar chart.
Opens a chart within a sandboxed iframe.
stack_components_analysisAnalyzes stack component usage.
Read-only analysis of stack component metadata.
recent_runs_analysisAnalyzes recent pipeline runs.
Read-only analysis of pipeline run metadata.
most_recent_runsRetrieves the N most recent pipeline runs.
Read-only access to pipeline run metadata.
API Key
hybrid
The ZenML MCP server is relatively safe for read-only operations. The greatest risks come from triggering pipelines and accessing secrets. Proper ZenML access controls and monitoring are essential to mitigate these risks.
The server primarily provides read-only access, limiting the scope for autonomous actions. Triggering pipelines requires careful consideration of resource consumption and potential side effects.
Production Tip
Monitor pipeline run activity and resource consumption to detect and prevent unauthorized or excessive pipeline triggering.
API Key or Service Account Token.
Use the `trigger_pipeline` tool with a snapshot name or ID.
No, only metadata about the artifacts is available.
Yes, they run in sandboxed iframes.
Configure resource limits within ZenML pipelines.
Snapshots are the preferred way to define runnable pipeline configurations. Run templates are deprecated.
Use Streamable HTTP transport and a publicly reachable URL (e.g., via Docker and Cloudflare Tunnel).