Conductor MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Conductor MCP Server Create a scheduled workflow to check stock prices daily."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Conductor mcp server
Model Context Protocol server for Conductor.
This package is used to run an MCP server that is capable of interacting with a Conductor instance. It provides tools for the basic operations that may be needed by an MCP client for Workflow creation, execution, and analysis.
PyPi Quickstart
Install package
pip install conductor-mcpRelated MCP server: ComfyUI MCP Server
Create a JSON config with your Conductor keys
{
"CONDUCTOR_SERVER_URL": "https://developer.orkescloud.com/api",
"CONDUCTOR_AUTH_KEY": "<YOUR_APPLICATION_AUTH_KEY>",
"CONDUCTOR_AUTH_SECRET": "<YOUR_APPLICATION_SECRET_KEY>"
}Note: the
/apipath is required as part of the CONDUCTOR_SERVER_URL for most applications
Plug the server into an AI Agent, such as Claude or Cursor
{
"mcpServers": {
"conductor": {
"command": "conductor-mcp",
"args": [
"--config",
"<ABSOLUTE PATH TO A JSON CONFIG FILE>"
]
}
}
}You should now be able to interact with Conductor via your AI Agent.
Adding to Claude
You can find instructions for adding to Claude here.
In general, you just add the mcpServers config (above) to your Claude config (or create it if it doesn't exist). For
instance, on Mac it might be ~/Library/Application\ Support/Claude/claude_desktop_config.json.
Adding to Cursor
The main Cursor instructions are here.
Go to Cursor -> Settings -> Cursor Settings -> MCP and select "+ Add new global MCP server".
Here you can add the exact same configuration file shown in the example for Claude (above). You can then access the AI chat feature and explore the MCP server in the sidebar with ⌘+L (Mac) or Ctrl+L (Windows/Linux).
Example prompts
Get Flight Risk Info
Create and execute a Conductor Workflow that calls any necessary http endpoints to gather current weather data around
Seattle and outputs the risk factors for flying a small airplane around the South Lake Union area using Visual Flight
Rules today. Only use publicly available endpoints that don't require an API key.Notify Stocks
(May require API Keys)
Create a Conductor Workflow that runs on a daily schedule, accepts a list of email address and a stock symbol, checks
current stock prices, and sends an email to everyone on the list if they should be happy or sad today based on stock
performance. Name the workflow "NotifyStonks" and use schemaVersion 2.GitHub Quickstart
Clone GitHub Repo
gh repo clone conductor-oss/conductor-mcpThis project relies on uv https://docs.astral.sh/uv/getting-started/
Create venv
(not entirely necessary, since uv automatically creates and uses the virtual environment on its own when running other commands)
uv sync
source .venv/bin/activateDefine Env Vars
You can continue to use a JSON config file and the --config flag, or if the server is running in an environment where
you have control over the environment variables the MCP server will look for them there if a config file is not
provided.
export CONDUCTOR_SERVER_URL="YOUR_CONDUCTOR_SERVER_URL"
export CONDUCTOR_AUTH_KEY="<YOUR_APPLICATION_AUTH_KEY>"
export CONDUCTOR_AUTH_SECRET="<YOUR_APPLICATION_SECRET_KEY>"Configure Your AI Assistant
{
"mcpServers": {
"conductor": {
"command": "uv",
"args": [
"--directory",
"<ABSOLUTE_PATH_TO_THE_PROJECT>",
"run",
"conductor-mcp",
"--config",
"<ABSOLUTE PATH TO A JSON CONFIG FILE>"
]
}
}
}Or Run Server Directly
cd <PROJECT_ROOT>
uv run conductor-mcp --config <ABSOLUTE PATH TO A JSON CONFIG FILE>Note: a
local_development.pyalso exists for setting env vars and will be used when the--local_devflag is set.
Available Tools
19 toolsevent_get_event_handlersA
Gets all event handlers or filters by event name and active status.
Event handlers define how Conductor responds to external events. They can trigger workflow executions or perform other actions when specific events occur.
Args: event: Optional event name to filter by active_only: If True, return only active event handlers (default: True)
Inputs: event, active_only. Returns: result
task_create_task_definitionA
Creates or updates a task definition.
The task definition should include at minimum:
name: The name/type of the task
description: A description of what the task does
retryCount: Number of retries (default: 3)
timeoutSeconds: Task timeout in seconds
responseTimeoutSeconds: Response timeout in seconds
Example task definition: { "name": "my_custom_task", "description": "A custom task that processes data", "retryCount": 3, "timeoutSeconds": 300, "responseTimeoutSeconds": 180, "inputKeys": ["input1", "input2"], "outputKeys": ["result"] }
Args: task_definition: A dictionary containing the task definition
Inputs: task_definition (required). Returns: result
task_get_all_task_definitionsA
Gets all task definitions
Inputs: no named parameters. Returns: result
task_get_task_by_idA
Gets the metadata for a conductor workflow task in json format based on that task's id
Args: task_id: The uuid representing the task id
Inputs: task_id (required). Returns: result
task_get_task_definition_for_tasktypeA
Gets the task definition for the given taskType.
"taskType" is synonymous with "task name".
This API refers to only user-defined tasks.Args: taskType: The string representing the desired tasks' taskType
Inputs: taskType (required). Returns: result
task_get_task_logsA
Gets execution logs for a specific task. Returns log entries generated during task execution.
Args: task_id: The uuid representing the task execution id
Inputs: task_id (required). Returns: result
task_get_task_queue_detailsA
Gets the current status details for all conductor workflow task queues
Inputs: no named parameters. Returns: result
task_update_task_statusA
Updates the status of a task execution. This is typically used by workers to update task status.
Args: task_id: The uuid representing the task execution id workflow_instance_id: The uuid representing the workflow instance id status: New task status (IN_PROGRESS, FAILED, FAILED_WITH_TERMINAL_ERROR, COMPLETED) output_data: Optional dictionary containing task output data logs: Optional list of log entries for the task
Inputs: task_id (required), workflow_instance_id (required), status (required), output_data, +1 more. Returns: result
workflow_create_workflow_definitionA
Creates a workflow definition from the provided workflow_definition dict param.
These are the main constructs that should be considered:
Do_while : to run through a list (input _items as the input parameters to iterate)
Switch: decision task
Inline: Executes javascript inline
In order to access workflow variables in javascript functions, they must first be assigned a value via inputParameters, for example: if there's a task named "analyze_market_data" then to access its output in javascript you need to create an inputParameter to assign it a value that can be used: "inputParameters": { "marketData": "${analyze_market_data.output}" } Then the "${analyze_market_data.output}" value can be accessed in javascript as "marketData" like so: "(function () {\n console.log("Market Data: " + $.marketData);\n})();"
In switch cases, an expression field must be defined using the same input method as above. You can construct the expression field like this: "inputParameters": { "marketData": "${analyze_market_data.output}" } "expression": "(function () {\n $.marketData.includes("NASDAQ");\n})();"
Here are some examples to help: ## Example of a switch task to run conditional tasks { "name": "switch_case_example", "description": "switch_case_example", "version": 1, "tasks": [ { "name": "switch", "taskReferenceName": "switch_ref", "inputParameters": { "switchCaseValue": "${workflow.input.case_param}" }, "type": "SWITCH", "decisionCases": { "case_a": [ { "name": "simple_1", "taskReferenceName": "simple_ref_1", "type": "SIMPLE" } ], "case_b": [ { "name": "simple", "taskReferenceName": "simple_ref", "type": "SIMPLE" } ] }, "defaultCase": [ { "name": "simple_2", "taskReferenceName": "simple_ref_2", "type": "SIMPLE" } ], "evaluatorType": "value-param" } ] }
## Example of a do_while for iterating over a list
{
"createTime": 1740724130693,
"name": "for_each_example",
"description": "for_each_example",
"version": 1,
"tasks": [
{
"name": "do_while",
"taskReferenceName": "do_while_ref",
"inputParameters": {
"items": "$ {workflow.input.items_list}"
},
"type": "DO_WHILE",
"loopOver": [
{
"name": "simple",
"taskReferenceName": "simple_ref",
"inputParameters": {
"item": "$ {do_while_ref.output.item}"
},
"type": "SIMPLE"
},
{
"name": "simple_1",
"taskReferenceName": "simple_ref_1",
"inputParameters": {
"item_to_process": "$ {do_while_ref.output.item}"
},
"type": "SIMPLE"
}
],
"evaluatorType": "value-param"
}
]
}
## Example of an inline javascript execution
{
"name": "inline_javascript_execution",
"version": 1,
"tasks": [
{
"name": "inline",
"taskReferenceName": "inline_ref",
"type": "INLINE",
"inputParameters": {
"invoker": "$ {workflow.input.invoker_name}"
},
"inputParameters": {
"expression": "(function () {\n return "Hello " + $.invoker;\n})();",
"evaluatorType": "graaljs",
"value1": 1,
"value2": 2
}
}
]
}
notice, that the javscript function is written as:
(function () {
return $.value1 + $.value2;
})();
and value1 and value2 are the input to the task.
## Important Rules
taskReferenceName MUST be unique in the workflow JSON.
When using INLINE task, all the variables in the script MUST be the input parameters to the task. Only task's input parameters can be accessed inside the script.
When trying to update a workflow that's already been created you must increment the version number, otherwise you need to pick a unique name for the workflow.
It's best not to use loopCondition, instead, iterate over the list of items using items input parameter. Nest SWITCH task if you want to do conditional processing.
### SWITCH task rules:
When using SWITCH task, the switchCaseValue _cannot_ contain expressions, scripts or methods. It has to be simple map. Use expression field to execute a script if required.
Remember, switchCaseValue and expression fields in SWITCH are mutually exclusive. ONLY one of them can be used. expression is NOT an input field, it is set as a property to the task.
expression MUST be a javascript function that returns a single string. Similar to INLINE task. The function is IIFE type, which is a JavaScript function that is defined and executed immediately.
When writing SWITCH task remember to consider all the cases, if required use defaultCase to handle default cases and the ones for which no clear branches are defined.
one more thing -- SWITCH task does not produce output, so do not use the output of a switch as an input to any task.
## Inline javascript rules
We use GraalVM to evaluate javascript code.
It's best not to use concat function to merge arrays or maps etc.
You can ONLY use the variables defined as input to the task in the javascript code. Access them as $.var.
Note, you _CANNOT_ use ${task.output.var} in javascript. neither in inline or Switch task expressions. Only $.var and var MUST be an input parameter to the task.
In order to address any inputs to the workflow or tasks, you must be sure to first assign that input to an input parameter, which then can be referenced -
for instance, in order to use ${workflow.input.case_param} in javascript, you must make sure the inputParameters assigns that to a variable in the task definition,
such as:
"inputParameters": {
"switchCaseValue": "${workflow.input.case_param}"
},
once that value is assigned to switchCaseValue via the inputParameters section of the task definition, it can be addressed in javascript as $.switchCaseValue
## Input mapping rules
* The task's input and output is always a Map data type. If a task's schema returns List or a single value as output, it is wrapped in a map with key "result"
* For HTTP task's the output is in "response" keyArgs: workflow_definition: A nested dictionary representing a workflow definition
Inputs: workflow_definition (required). Returns: result
workflow_get_all_workflowsA
Gets a short description of all existing conductor workflows.
Inputs: no named parameters. Returns: result
workflow_get_workflow_by_idB
Gets a conductor workflow execution in json format based on the workflow's execution id
Args: workflow_id: The uuid representing the execution of the workflow
Inputs: workflow_id (required). Returns: result
workflow_get_workflow_by_nameB
Gets the metadata for a conductor workflow in json format based on that workflow's name
Args: workflow_name: The name of the workflow
Inputs: workflow_name (required). Returns: result
workflow_pause_workflowA
Pauses a running workflow execution. The workflow will pause and can be resumed later.
Args: workflow_id: The uuid representing the execution of the workflow to pause
Inputs: workflow_id (required). Returns: result
workflow_query_workflow_executionsA
Search for workflow (executions) based on payload and other parameters. The query parameter accepts exact matches using = and AND operators on the following fields: workflowId, correlationId, workflowType, and status. Matches using = can be written as taskType = HTTP. Matches using IN are written as status IN (SCHEDULED, IN_PROGRESS). The 'startTime' and 'modifiedTime' field uses unix timestamps and accepts queries using < and >, for example startTime < 1696143600000. Queries can be combined using AND, for example taskType = HTTP AND status = SCHEDULED
If no query kwargs are provided, all workflow executions will be returned.
Example call to this function to query for a status of FAILED and start time after Thu May 01 2025 22:20:59 GMT+0000: query_workflow_executions('status="FAILED" AND startTime > 1746138025 ')
Searching for a range of time does not work, i.e. "startTime > 0 AND startTime < 1746138025"
Example call for FAILED or COMPLETED status and workflow named "SimpleWorkflow": query_workflow_executions('status IN (FAILED, COMPLETED) AND workflowType="SimpleWorkflow"')
Args: query: A query string, utilizing any of the following fields. workflowId: The id of a workflow execution. correlationId: The correlationId used to create any workflow executions. workflowType: Synonymous with workflow name. createTime: The creation unix timestamp of a workflow. startTime: The start unix timestamp of a workflow. status: The status of a workflow execution. One of [RUNNING, PAUSED, COMPLETED, TIMED_OUT, TERMINATED, FAILED]. endTime: The end unix timestamp of a workflow.
Inputs: query (required). Returns: result
workflow_restart_workflowA
Restarts a workflow execution from the beginning. This creates a new execution with the same input.
Args: workflow_id: The uuid representing the execution of the workflow to restart use_latest_definitions: If True, use the latest workflow definition instead of the original version
Inputs: workflow_id (required), use_latest_definitions. Returns: result
workflow_resume_workflowA
Resumes a paused workflow execution. The workflow will continue from where it was paused.
Args: workflow_id: The uuid representing the execution of the workflow to resume
Inputs: workflow_id (required). Returns: result
workflow_retry_workflowB
Retries a failed workflow execution from the last failed task.
Args: workflow_id: The uuid representing the execution of the workflow to retry resume_subworkflow_tasks: If True, resume any subworkflow tasks that were running
Inputs: workflow_id (required), resume_subworkflow_tasks. Returns: result
workflow_start_workflow_by_nameA
Starts a new execution of a conductor workflow by its name
Args: workflow_name: The name of the workflow definition to create a new execution for correlation_id: An integer used as unique identifier for the workflow execution, used to correlate the current workflow instance with other workflows. priority: A number starting at 0 representing the priority of the execution of the workflow. Lower numbers mean higher priority. idempotency_key: An arbitrary, user-provided string used to ensure idempotency when calling this endpoint multiple times. idempotency_strategy: A string representing one of the following three strategies: RETURN_EXISTING: Return the workflowId of the workflow instance with the same idempotency key. FAIL: Start a new workflow instance only if there are no workflow executions with the same idempotency key. FAIL_ON_RUNNING: Start a new workflow instance only if there are no RUNNING or PAUSED workflows with the same idempotency key. Completed workflows can run again. data: A dictionary containing any arguments to pass into the workflow for creation
Inputs: workflow_name (required), correlation_id, priority, idempotency_strategy, +2 more. Returns: result
workflow_terminate_workflowA
Terminates a workflow execution. This will stop the workflow and mark it as terminated.
Args: workflow_id: The uuid representing the execution of the workflow to terminate reason: Optional reason for termination
Inputs: workflow_id (required), reason. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 19 tools
Tools are clearly grouped by resource (event, task, workflow) and each tool has a distinct purpose. There is no overlap between tools, and descriptions make it easy to differentiate them.
All tool names follow a consistent resource_verb_noun pattern (e.g., event_get_event_handlers, task_create_task_definition, workflow_get_workflow_by_id). This provides predictability and clarity.
With 19 tools, the set is slightly above the typical 3-15 range but still reasonable for a comprehensive workflow orchestration server covering event handling, task management, and workflow lifecycle operations.
The tool surface covers major operations for event handlers, task definitions, and workflow executions. Minor gaps exist, such as explicit workflow definition update and task execution result retrieval, but the overall coverage is good.
Maintenance
Related MCP Connectors
Build, test and publish agentic workflows from your AI client: approve steps, read tables, see runs.
Build and run agents and automations across hundreds of apps using natural language.
Create and manage AI agents that collaborate and solve problems through natural language interacti…
The AI orchestration agent for modern software teams.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agent and task management using the CrewAI framework, allowing users to create and run agents and tasks in an automated workflow environment.6Apache 2.0
- AlicenseAqualityDmaintenanceEnables AI agents to manage ComfyUI workflows using a human-readable Domain Specific Language (DSL), with automatic conversion to/from JSON format. Supports workflow creation, validation, execution, and monitoring through natural language interactions.163MIT
- AlicenseBqualityNot gradedmaintenanceEnables AI assistants to interact with Netflix Conductor Workflow Engine for troubleshooting workflows, managing executions, and creating/updating workflow and task definitions through natural language.192-
- AlicenseBqualityBmaintenanceEnables AI assistants to manage Temporal workflows, schedules, and workflow executions through a standardized interface.313,001 PyPI66Apache 2.0