---
title: "Test Automation"
description: "Run an automation once against one media item to see which way it branches."
---

> Documentation Index
> Fetch the complete documentation index at: https://docs.speakai.co/llms.txt
> Use this file to discover all available pages before exploring further.

# Test Automation

Use the Speak AI MCP server tool `test_automation` to run an automation once against
one media item to see which way it branches.

## What does it do?

Run an automation once against one media item to see which way it branches. THIS HAS
REAL SIDE EFFECTS: only Speak's own run notifications are suppressed. Outbound
webhooks fire, Composio actions run against the connected third party, and AI steps
are billed. Ask the user before calling it on an automation that posts anywhere
outside Speak. It also needs a mediaId and sends no webhook payload, so an
inbound-webhook automation cannot be meaningfully tested this way: its payload
tokens will resolve to empty. Returns a runId; read the result with
`get_automation_run`.

## Parameters

`test_automation` takes 2 parameters, all of them required.

| Name | Type | Required | Description |
| --- | --- | --- | --- |
| `automationId` | string | Yes | Unique identifier of the automation to test |
| `mediaId` | string | Yes | Media item to run the automation against |

## Behaviour

The Speak AI MCP server publishes these behaviour hints for `test_automation`.

- Can change data in your workspace.
- Can delete or overwrite data. Use with care.
- Repeating an identical call can have an extra effect.
- Can reach systems outside Speak AI.

## Related

- [Automations](/mcp/tools/automations/) lists the rest of the Automations category.
- [Tool reference](/mcp/tools/) lists every Speak AI MCP server tool.
- [MCP server](/mcp/) covers setup, supported clients, and what agents can do once
  connected.

Source: https://docs.speakai.co/mcp/tools/automations/test_automation/index.mdx