Getting Started with Informatica MCP Servers > Shared MCP servers > Job Management server
  

Job Management server

The Job Management server provides the operational interface between your AI agents and the IDMC execution engine. It allows agents to orchestrate data pipelines, monitor process health, and manage task lifecycles in real time.
The Job Management server provides the operational interface between your AI agents and the IDMC execution engine. It allows agents to orchestrate data pipelines, monitor process health, and manage task lifecycles in real time. By exposing Data Integration controls, this MCP server enables your AI agent to act as an automated operator, ensuring that jobs are triggered, tracked, and managed without manual intervention.

Key capabilities

This MCP server offers the following key capabilities:

Use cases

Use this MCP server to address the following use cases:

Tools

The following table describes the tools available on this MCP server:
Tools
Description
get_dynamic_mapping_task_status
Retrieves run status, row counts, and job execution details for a dynamic mapping task run.
get_job_status
Retrieves real-time execution states and operational statistics, including row counts and error logs, for a specific job instance.
get_taskflow_status
Retrieves the run status of published taskflows from the Data Integration taskflow runtime. All parameters are optional. Call with no parameters to list recent taskflow runs or to narrow the results by run ID, run status, or a start/end window.
list_running_jobs
Lists Data Integration jobs that are currently running or queued, as reported by Monitor. Returns a summary for each job. Set the details to true to include subtask entries for each job.
resume_taskflow_fault_entry
Resumes a suspended taskflow by retrying the step that faulted, identified by run ID with an empty request body. This is not applicable to taskflows configured with custom error handling. This tool changes system state by resuming the referenced job.
resume_taskflow_fault_skip
Resumes a suspended taskflow by skipping the step that faulted and continuing with the next step. Identified by run ID with an empty request body. This changes system state by resuming the referenced job.
run_dynamic_mapping_task
Runs a dynamic mapping task identified by its federated task ID.
run_linear_taskflow
Runs a linear taskflow (a sequenced set of Data Integration tasks) identified by its task ID, task name, or federated task ID.
run_mapping_task
Starts job execution by initiating a specific mapping task with the required runtime parameters.
run_masking_task
Runs a data masking task, identified by its task ID, task name, or federated task ID to protect sensitive data.
run_powercenter_task
Runs a PowerCenter task in Data Integration, identified by its task ID, task name, or federated task ID.
run_published_taskflow
Runs a published taskflow, identified by its Run API name, and returns the new execution's run identifier. The request body carries the taskflow's input fields, which vary per taskflow.
Note:
The run identifier is returned in the capitalized RunId field inside RunTaskFlowResponse. Downstream taskflow tools such as get_taskflow_status, resume_taskflow_fault_retry, resume_taskflow_fault_skip, and terminate_taskflow_runs expect the same value in the lowercase runId or runid field. Callers must map RunId to runid when chaining a run into a status, resume, or terminate call.
run_replication_task
Runs a data replication task identified by its task ID, task name, or federated task ID that includes the folder path.
run_synchronization_task
Runs a data synchronization task that keeps a source and target in sync, identified by its task ID, task name, or federated task ID.
search_completed_jobs
Returns completed Data Integration jobs with run execution details, completion state, source and target row counts, error messages, and transformation statistics.
All parameters are optional. Call with no parameters to search all completed jobs. Supply ID searches for a specific log entry. Task ID searches for a specific task. When used together, pass ID and run ID searches for one completed job. Use offset and rowLimit to search through large result sets.
search_task_asset
Searches Data Integration tasks by type, then returns the associated federated IDs to use as the taskFederatedId when you submit a job request.
stop_dynamic_mapping_task
Stops a running dynamic mapping task by submitting its federated task ID and run instance ID.
stop_running_job
Sends an immediate termination request to the Secure Agent to stop the execution of an active job.
terminate_taskflow_runs
Terminates one or more in-progress taskflow jobs, identified by their run IDs, with a maximum of 200 per request.
Note:
In AI Agent Engineering, the MCP server tools are called "actions."