Test run automation MCP tool

Start a manual test run for an existing GovTribe AI automation and return the started conversation, including a GovTribe URL when a conversation is available.

MCP tool name: Test_Run_Automation
Returns: Test run automation MCP response
Annotations: Not read only, destructive, not idempotent, closed world
Surfaces: GovTribe AI

Examples

Use this tool to start a manual run against an existing automation. When the automation has multiple start nodes, send the stable start_node_id for the condition the test should simulate.

Load the available start nodes

{
  "tool": "Search_Automations",
  "arguments": {
    "automation_workflow_ids": [
      "<automation_govtribe_id>"
    ],
    "fields_to_return": [
      "govtribe_id",
      "name",
      "start_nodes",
      "total_triggers_count"
    ],
    "max_triggers": 50,
    "past_runs_limit": 0
  }
}

If total_triggers_count is greater than the number of returned start_nodes, repeat the search with a larger max_triggers before choosing a start node.

Simulate one condition on a multi-trigger automation

{
  "tool": "Test_Run_Automation",
  "arguments": {
    "workflow_id": "<automation_govtribe_id>",
    "start_node_id": "<government_record_start_node_id>"
  }
}

For a government-record trigger, GovTribe uses recent matching activity when it is available. Otherwise, the manual run uses clearly synthetic test activity and does not create a government-record change.

Start a single-trigger scheduled automation

start_node_id can be omitted only when the automation has one start node.

{
  "tool": "Test_Run_Automation",
  "arguments": {
    "workflow_id": "<automation_govtribe_id>"
  }
}

Start a saved-search test run with only five result rows

{
  "tool": "Test_Run_Automation",
  "arguments": {
    "workflow_id": "<automation_govtribe_id>",
    "start_node_id": "<saved_search_start_node_id>",
    "saved_search_result_limit": 5
  }
}

Start a stage-change test run with explicit context

{
  "tool": "Test_Run_Automation",
  "arguments": {
    "workflow_id": "<automation_govtribe_id>",
    "start_node_id": "<stage_change_start_node_id>",
    "pipeline_id": "<pipeline_govtribe_id>",
    "stage_id": "<stage_govtribe_id>"
  }
}

Payload patterns

Always send workflow_id.

Argument	When to send it
`start_node_id`	Required when the automation has more than one start node. For a single-start automation, omit it or send that node's stable ID.
`pursuit_id`	Add when a pursuit-based trigger needs pursuit context for the test.
`pipeline_id`	Add when a pipeline-based trigger needs pipeline context for the test.
`stage_id`	Add when a stage-change trigger is not already scoped to a specific stage.
`saved_search_result_limit`	Add for a saved-search test when you want to limit included matches. Values can range from 1 to 25.

{
  "tool": "Test_Run_Automation",
  "arguments": {
    "workflow_id": "<automation_govtribe_id>",
    "start_node_id": "<selected_start_node_id>"
  }
}

Test run automation MCP response: Review the response shape returned by this tool.
GovTribe AI: Review the GovTribe AI surface where this tool can be called.
Source identifiers and record matching: Use GovTribe IDs and source identifiers when passing records between tools.

On this page