mcp/testkube

Verified Publisher

By mcp

•Updated 14 days ago

The Testkube MCP Server exposes continuous testing capabilities (test orchestration, execution, t...

Image

Machine learning & AI

10K+

View on MCP:Hub

Overview Tags

mcp/testkube repository overview

⁠Testkube MCP Server

The Testkube MCP Server exposes continuous testing capabilities (test orchestration, execution, troubleshooting and anlysis) to AI tools and workflows .

What is an MCP Server?⁠

⁠MCP Info

Attribute	Details
Docker Image	mcp/testkube⁠
Author	kubeshop⁠
Repository	https://github.com/kubeshop/testkube⁠

⁠Image Building Info

Attribute	Details
Dockerfile	https://github.com/kubeshop/testkube/blob/d392deb5bf6876064b2499f92fd78308aeb57c45/build/mcp-server/Dockerfile⁠
Commit	`d392deb5bf6876064b2499f92fd78308aeb57c45`
Docker Image built by	Docker Inc.
Docker Scout Health Score
Verify Signature	`COSIGN_REPOSITORY=mcp/signatures cosign verify mcp/testkube --key https://raw.githubusercontent.com/docker/keyring/refs/heads/main/public/mcp/latest.pub`
Licence	Other

⁠Available Tools (28)

Tools provided by this Server	Short Description
`abort_workflow_execution`	Abort a running workflow execution.
`build_dashboard_url`	Build dashboard URLs for Testkube workflows and executions.
`create_workflow`	Create a new TestWorkflow from a YAML definition.
`create_workflowtemplate`	Create a new TestWorkflowTemplate from a YAML definition.
`fetch_execution_logs`	Retrieve logs from a test workflow execution.
`get_execution_info`	Get detailed information about a specific workflow execution: status, timing, results, configuration, and worker instances.
`get_execution_schema`	Get the YAML schema for TestWorkflowExecution data.
`get_workflow`	Get detailed workflow information including execution history, health metrics, and current status.
`get_workflow_definition`	Get the YAML definition of a specific Testkube workflow.
`get_workflow_execution_metrics`	Get raw resource metrics (CPU, memory, disk, network) for a single workflow execution as time-series data.
`get_workflow_metrics`	Get execution metrics for a workflow: execution statistics, health scores, pass rates, and performance data.
`get_workflow_resource_history`	Analyze resource consumption patterns (CPU, memory, disk, network) across recent executions of a workflow.
`get_workflow_schema`	Get the YAML schema for TestWorkflow definitions.
`get_workflowtemplate_definition`	Get the YAML definition of a specific TestWorkflowTemplate.
`list_agents`	List available agents in the organization for workflow execution targeting.
`list_artifacts`	List all artifacts (files, reports, logs) generated by a workflow execution.
`list_executions`	List test workflow executions with filtering by workflow, status, date range, labels, tags, or text search.
`list_labels`	List all workflow labels and their values in the environment.
`list_resource_groups`	List available resource groups in the organization.
`list_workflows`	List Testkube workflows with optional filtering by resource group, selector, status, and text search.
`list_workflowtemplates`	List all TestWorkflowTemplates with optional label filtering.
`lookup_execution_id`	Resolve an execution name (e.g., 'my-workflow-123') to its execution ID.
`read_artifact`	Read content from an artifact file produced by a workflow execution.
`run_workflow`	Run a TestWorkflow with optional configuration parameters and agent targeting.
`update_execution_tags`	Update tags on a workflow execution.
`update_workflow`	Update an existing TestWorkflow with a new YAML definition.
`update_workflowtemplate`	Update an existing TestWorkflowTemplate with a new YAML definition.
`wait_for_executions`	Wait for a list of workflow executions to complete.

⁠Tools Details

⁠Tool: `abort_workflow_execution`

Abort a running workflow execution. Stops the execution and marks it as aborted. Use for cancelling long-running or stuck executions.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `build_dashboard_url`

Build dashboard URLs for Testkube workflows and executions. Supports deep linking to a specific step in the execution log view via stepRef.

Parameters	Type	Description
`resourceType`	`string`	Type of dashboard resource: 'workflow' or 'execution'
`workflowName`	`string`	Name of the test workflow
`executionId`	`string` optional	Execution ID (required for execution URLs)
`executionTab`	`string` optional	Execution tab to navigate to (e.g. 'log-output'). Defaults to 'log-output' when stepRef is provided.
`stepRef`	`string` optional	Step reference ID to deep link to a specific workflow step in the log view (e.g. 'rwhc2zn'). Obtain from execution info signatures. Requires executionId.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `create_workflow`

Create a new TestWorkflow from a YAML definition. The workflow is immediately available for execution after creation.

Parameters	Type	Description
`yaml`	`string`	Complete YAML definition of the TestWorkflow to create in Testkube. This should be the full workflow specification including metadata, spec, and all steps.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `create_workflowtemplate`

Create a new TestWorkflowTemplate from a YAML definition. The template is immediately available for use by workflows.

Parameters	Type	Description
`yaml`	`string`	Complete YAML definition of the TestWorkflowTemplate to create in Testkube. This should be the full template specification including metadata and spec.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `fetch_execution_logs`

Retrieve logs from a test workflow execution. Default returns last 100 lines. Use grep to search the full log (capped at 100 matches). Always paginate in 100-line chunks. For parallel workflows with workers, call get_execution_info first to get valid worker refs (do not use step refs).

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB format (e.g., '67d2cdbc351aecb2720afdf2').
`endLine`	`string` optional	1-based line number to stop reading at (inclusive). Use with startLine for a range.
`grep`	`string` optional	Filter to lines containing this substring (e.g., grep=ERROR).
`startLine`	`string` optional	1-based line number to start reading from. Use with endLine for a range.
`step`	`string` optional	Filter to logs from a specific workflow step by reference name (e.g., 'run-tests', 'setup-env').
`tail`	`string` optional	Return the last N lines of the log. Example: '50'
`workerIndex`	`string` optional	0-based index of the parallel worker to fetch logs from (default: 0). Use with workerRef.
`workerRef`	`string` optional	Worker instance ref from the 'workers' array returned by get_execution_info (e.g., 'r72qph9'). ONLY use values from that array — do NOT use step refs from the main log metadata. When set, fetches that specific worker's logs instead of the main execution log.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_execution_info`

Get detailed information about a specific workflow execution: status, timing, results, configuration, and worker instances. Requires executionId. workflowName is optional for disambiguation.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.
`workflowName`	`string` optional	Optional workflow name for scoping an execution name lookup. Safe to omit when you have an execution ID.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_execution_schema`

Get the YAML schema for TestWorkflowExecution data. Returns all available fields, their types, and descriptions. Use to understand execution data structure when analyzing results.

⁠Tool: `get_workflow`

Get detailed workflow information including execution history, health metrics, and current status.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_workflow_definition`

Get the YAML definition of a specific Testkube workflow. Returns the complete specification including steps, configuration schema, and metadata.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_workflow_execution_metrics`

Get raw resource metrics (CPU, memory, disk, network) for a single workflow execution as time-series data. Use for deep-dive debugging of a specific run. Requires workflowName and executionId.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.
`maxSamples`	`number` optional	Maximum number of time-series data points to return per metric (default: 50). Increase for more granular data, decrease for a more compact response.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_workflow_metrics`

Get execution metrics for a workflow: execution statistics, health scores, pass rates, and performance data. Use to analyze workflow reliability over time.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_workflow_resource_history`

Analyze resource consumption patterns (CPU, memory, disk, network) across recent executions of a workflow. Computes cross-execution statistics (mean, min, max, stdDev), detects trends, and identifies outliers. Use to investigate growing resource usage or find abnormal runs. Requires workflowName.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.
`lastN`	`string` optional	Number of recent executions to analyze (default: 50, max: 100)
`metrics`	`string` optional	Filter to specific metrics: cpu, memory, disk, network. Comma-separated for multiple. Default: all metrics.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `get_workflow_schema`

Get the YAML schema for TestWorkflow definitions. Returns all available fields, their types, and descriptions. Use to understand workflow structure when creating or querying workflows.

⁠Tool: `get_workflowtemplate_definition`

Get the YAML definition of a specific TestWorkflowTemplate. Returns the complete template specification including steps, config schema, and metadata.

Parameters	Type	Description
`templateName`	`string`	The name of the workflow template. Template names are lowercase alphanumeric with dashes
(e.g., 'my-template', 'official--k6--v1'). This uniquely identifies a TestWorkflowTemplate within the environment.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `list_agents`

List available agents in the organization for workflow execution targeting. Returns agent names, types, capabilities, labels, and status. Use before run_workflow to discover target agents.

Parameters	Type	Description
`capability`	`string` optional	Filter by capability (e.g., 'runner')
`page`	`string` optional	Page number (default: 0)
`pageSize`	`string` optional	Number of items per page (default: 20)
`type`	`string` optional	Filter by agent type (e.g., 'runner')

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `list_artifacts`

List all artifacts (files, reports, logs) generated by a workflow execution. Returns artifact names, sizes, and status. Use to discover available outputs before reading specific artifacts.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `list_executions`

List test workflow executions with filtering by workflow, status, date range, labels, tags, or text search. Returns execution summaries including status, duration, and metadata. Use to discover recent runs or find specific executions.

Parameters	Type	Description
`endDate`	`string` optional	Filter items on or before this time. Accepts a date (YYYY-MM-DD, e.g., '2024-01-31')
or an RFC 3339 timestamp (e.g., '2024-01-31T16:00:00Z'). Combine with startDate for date ranges.
`page`	`string` optional	Page number for pagination (default: 0)
`pageSize`	`string` optional	Number of items to return per page (default: 10, max: 100)
`selector`	`string` optional	Filter workflows by label using key=value format. For single label use 'key=value',
for multiple labels use comma-separated format 'key1=value1,key2=value2' (e.g., 'tool=cypress,env=prod').
Note: filters workflow-level labels, not execution tags — use tagSelector for execution tags.
`since`	`string` optional	Filter executions created after this time (ISO 8601 format)
`startDate`	`string` optional	Filter items on or after this time. Accepts a date (YYYY-MM-DD, e.g., '2024-01-15')
or an RFC 3339 timestamp (e.g., '2024-01-15T13:00:00Z').
`status`	`string` optional	Filter by execution status. Available statuses: 'queued', 'running', 'passed',
'failed', 'skipped', 'aborted', 'timeout', 'paused'
`tagSelector`	`string` optional	Filter executions by tag using key=value format. For single tag use 'key=value',
for multiple tags use comma-separated format 'key1=value1,key2=value2' (e.g., 'type=suite,env=prod').
Note: filters execution-level tags (set via update_execution_tags), not workflow labels — use selector for workflow labels.
`textSearch`	`string` optional	Text search filter for names or descriptions. Can use space-separated words
to find items containing all terms
`workflowName`	`string` optional	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `list_labels`

List all workflow labels and their values in the environment. Use to discover available filters for selector parameters in other tools.

⁠Tool: `list_resource_groups`

List available resource groups in the organization. Use to discover group slugs for filtering workflows and executions.

⁠Tool: `list_workflows`

List Testkube workflows with optional filtering by resource group, selector, status, and text search. Returns workflow names, descriptions, and execution status.

Parameters	Type	Description
`page`	`string` optional	Page number for pagination (default: 0)
`pageSize`	`string` optional	Number of items to return per page (default: 10, max: 100)
`resourceGroup`	`string` optional	Filter by resource group using the group slug (e.g., 'demo-resource-group', 'accounting-tests'). Use the list_resource_groups tool to discover available groups
`selector`	`string` optional	Filter workflows by label using key=value format. For single label use 'key=value',
for multiple labels use comma-separated format 'key1=value1,key2=value2' (e.g., 'tool=cypress,env=prod').
Note: filters workflow-level labels, not execution tags — use tagSelector for execution tags.
`status`	`string` optional	Filter by execution status. Available statuses: 'queued', 'running', 'passed',
'failed', 'skipped', 'aborted', 'timeout', 'paused'
`textSearch`	`string` optional	Text search filter for names or descriptions. Can use space-separated words
to find items containing all terms

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `list_workflowtemplates`

List all TestWorkflowTemplates with optional label filtering. Returns template names, descriptions, and labels.

Parameters	Type	Description
`selector`	`string` optional	Filter workflows by label using key=value format. For single label use 'key=value',
for multiple labels use comma-separated format 'key1=value1,key2=value2' (e.g., 'tool=cypress,env=prod').
Note: filters workflow-level labels, not execution tags — use tagSelector for execution tags.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `lookup_execution_id`

Resolve an execution name (e.g., 'my-workflow-123') to its execution ID. Use when you have an execution name but need the ID for other tools.

Parameters	Type	Description
`executionName`	`string`	The name of the execution (e.g., 'my-workflow-123'). Execution names follow
the pattern of workflow name plus a numeric suffix. Use lookup_execution_id to get the ID from a name.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `read_artifact`

Read content from an artifact file produced by a workflow execution. Default returns first 100 lines (max 200 per request). Always paginate in 100-200 line chunks. For binary artifacts, returns a summary instead of content.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.
`fileName`	`string`	The name of the artifact file to retrieve
`endLine`	`string` optional	1-based line number to stop reading at (inclusive). Use with startLine for a range.
`grep`	`string` optional	Case-insensitive substring filter. Returns matching lines with 3 lines of context. When set, startLine and endLine are ignored — use grep OR line range, not both.
`startLine`	`string` optional	1-based line number to start reading from. Use with endLine for a range.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `run_workflow`

Run a TestWorkflow with optional configuration parameters and agent targeting. Use get_workflow_definition first to discover available config parameters. Use list_agents to discover available target agents.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.
`config`	`object` optional	Configuration parameters for the workflow. Use get_workflow_definition tool first to examine the spec.config section to see what parameters are available.
`target`	`object` optional	Target specification for multi-agent execution (optional). Supports: {"name": "agent-name"} for name-based targeting, {"labels": {"env": "prod"}} for label-based targeting, or standard ExecutionTarget format with match/not/replicate fields.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `update_execution_tags`

Update tags on a workflow execution. Uses replace semantics: provided tags completely replace existing tags. Send empty map {} to clear all tags. Tags are key-value pairs for categorization and filtering.

Parameters	Type	Description
`executionId`	`string`	The unique execution ID in MongoDB ObjectID format (24 hex chars, e.g., '67d2cdbc351aecb2720afdf2').
Use lookup_execution_id if you only have an execution name.
`tags`	`object`	Key-value tag pairs (e.g., {"env":"prod","bug":"found"}). Replaces all existing tags. Use {} to clear all tags.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `update_workflow`

Update an existing TestWorkflow with a new YAML definition. The workflow is updated immediately and available for execution with the new configuration.

Parameters	Type	Description
`workflowName`	`string`	The name of the workflow. Workflow names are lowercase alphanumeric with dashes
(e.g., 'my-workflow', 'api-tests'). This uniquely identifies a TestWorkflow within the organization.
`yaml`	`string`	Complete YAML definition of the TestWorkflow to update in Testkube. This should be the full workflow specification including metadata, spec, and all steps.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `update_workflowtemplate`

Update an existing TestWorkflowTemplate with a new YAML definition. Workflows using the template pick up the changes.

Parameters	Type	Description
`templateName`	`string`	The name of the workflow template. Template names are lowercase alphanumeric with dashes
(e.g., 'my-template', 'official--k6--v1'). This uniquely identifies a TestWorkflowTemplate within the environment.
`yaml`	`string`	Complete YAML definition of the TestWorkflowTemplate to update in Testkube. This should be the full template specification including metadata and spec.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Tool: `wait_for_executions`

Wait for a list of workflow executions to complete. Returns the final status of all executions. Use for synchronizing dependent workflows.

Parameters	Type	Description
`executionIds`	`string`	Comma-separated list of execution IDs to wait for (e.g., 'exec1,exec2,exec3').
`timeoutMinutes`	`string` optional	Maximum time to wait in minutes before timing out (default: 30). Set to 0 for no timeout.

This tool may perform destructive updates.

This tool interacts with external entities.

⁠Use this MCP Server

{
  "mcpServers": {
    "testkube": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_CONTROL_PLANE_URL",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "mcp/testkube"
      ],
      "env": {
        "TK_CONTROL_PLANE_URL": "https://api.testkube.io",
        "TK_ACCESS_TOKEN": "tkcapi_xxxxxxx",
        "TK_ORG_ID": "tkcorg_xxxxxxx",
        "TK_ENV_ID": "tkcenv_xxxxxxx"
      }
    }
  }
}

Why is it safer to run MCP Servers with Docker?⁠

Install from MCP:Hub

Testkube

View and Install from MCP:Hub

Tag summary

Recent tags

Content type

Image

Digest

sha256:43aa88f2d…

Size

44.4 MB

Last updated

14 days ago

docker pull mcp/testkube

This week's pulls

Pulls:

948

Jul 6 to Jul 12

Learn more⁠

mcp/testkube

mcp/testkube repository overview

⁠Testkube MCP Server

⁠MCP Info

⁠Image Building Info

⁠Available Tools (28)

⁠Tools Details

⁠Tool: abort_workflow_execution

⁠Tool: build_dashboard_url

⁠Tool: create_workflow

⁠Tool: create_workflowtemplate

⁠Tool: fetch_execution_logs

⁠Tool: get_execution_info

⁠Tool: get_execution_schema

⁠Tool: get_workflow

⁠Tool: get_workflow_definition

⁠Tool: get_workflow_execution_metrics

⁠Tool: get_workflow_metrics

⁠Tool: get_workflow_resource_history

⁠Tool: get_workflow_schema

⁠Tool: get_workflowtemplate_definition

⁠Tool: list_agents

⁠Tool: list_artifacts

⁠Tool: list_executions

⁠Tool: list_labels

⁠Tool: list_resource_groups

⁠Tool: list_workflows

⁠Tool: list_workflowtemplates

⁠Tool: lookup_execution_id

⁠Tool: read_artifact

⁠Tool: run_workflow

⁠Tool: update_execution_tags

⁠Tool: update_workflow

⁠Tool: update_workflowtemplate

⁠Tool: wait_for_executions

⁠Use this MCP Server

Install from MCP:Hub

Testkube

Tag summary

This week's pulls

⁠Tool: `abort_workflow_execution`

⁠Tool: `build_dashboard_url`

⁠Tool: `create_workflow`

⁠Tool: `create_workflowtemplate`

⁠Tool: `fetch_execution_logs`

⁠Tool: `get_execution_info`

⁠Tool: `get_execution_schema`

⁠Tool: `get_workflow`

⁠Tool: `get_workflow_definition`

⁠Tool: `get_workflow_execution_metrics`

⁠Tool: `get_workflow_metrics`

⁠Tool: `get_workflow_resource_history`

⁠Tool: `get_workflow_schema`

⁠Tool: `get_workflowtemplate_definition`

⁠Tool: `list_agents`

⁠Tool: `list_artifacts`

⁠Tool: `list_executions`

⁠Tool: `list_labels`

⁠Tool: `list_resource_groups`

⁠Tool: `list_workflows`

⁠Tool: `list_workflowtemplates`

⁠Tool: `lookup_execution_id`

⁠Tool: `read_artifact`

⁠Tool: `run_workflow`

⁠Tool: `update_execution_tags`

⁠Tool: `update_workflow`

⁠Tool: `update_workflowtemplate`

⁠Tool: `wait_for_executions`