> For the complete documentation index, see [llms.txt](https://help.smartico.ai/welcome/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-build-and-copy-campaigns-and-journeys.md).

# MCP - Build and copy campaigns and journeys

With the [MCP Server](https://help.smartico.ai/welcome/technical-guides/mcp-server), an AI assistant can read a campaign and its journey, build a new one step by step, clone it, and rebuild it on another label. Every change can be checked first with an optional dry run, and the assistant asks you before it starts, stops or changes the speed of a campaign.

## How a campaign is built

A campaign has two parts, and MCP edits them with different tools:

| Part               | What it holds                                                                                                                                       | Read with                                       | Change with                                                                                              |
| ------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------- | -------------------------------------------------------------------------------------------------------- |
| **Entry settings** | Name, realtime or scheduled, trigger event and entry conditions, target segment, schedule, entry mode, control group, limits, status                | `entity_get` (type `campaign`)                  | `entity_create`, `entity_update`                                                                         |
| **Journey (flow)** | The steps players go through after they enter: communications, delays, points, bonuses, wait-for-event steps, splits, and the branches between them | `read_operation` → `flow_get`, `flow_get_nodes` | `exec_operation` → `flow_add_node`, `flow_update_node`, `flow_set_edge`, `flow_copy_node`, `flow_layout` |

A new campaign is always created as a **Draft**. It runs only after you make it **Active**.

Some entry settings are fixed when the campaign is created: a campaign cannot be switched between realtime and scheduled, and the entry mode "Once in a lifetime of a user" cannot be changed later. A schedule set at one of the busiest times of the day is saved, and the answer adds a warning that names a quieter time.

## Read a journey

Ask the assistant to read a campaign, for example *"Show me the journey of campaign 'Welcome series' as a diagram."* The `flow_get` operation returns the journey as an outline: one row per step, with its parent, its branch and a short label, but without the step's settings. With `mermaid: true` it also returns a diagram that the assistant can draw. To see the settings of steps, `flow_get_nodes` reads up to 20 steps in full. Both are read-only, so the assistant can run them without asking.

Also read-only: `flow_describe_activity` (the fields of a step type), `list_stop_convert_rules` (the stop and convert rules) and `list_campaign_tags` (the campaign's shared personalization values).

## Add and change steps

| Operation                                           | What it does                                                                                                                                                                                                                                                                                                                                                                                                                                   |
| --------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `flow_add_node`                                     | Adds a step at the start of the journey or on a branch of another step: a communication (email, push, popup, inbox and other channels), a delay, points, a bonus, a wait-for-event step or a split. It can also add a whole branch of steps in one call (`children`, up to 60 steps), put a new step above an existing one (`insert_before`), or add a merge: an arrow from a branch into an existing step, so two paths continue in one step. |
| `flow_update_node`                                  | Changes a step's settings, moves it, or changes its status (active, paused, disabled, archived). Archiving a step removes it and the steps under it. For a step that a merge leads into, it also sets whether the step runs again when a player arrives from the second path ("Run this step again").                                                                                                                                          |
| `flow_set_edge`                                     | Changes the edge that leads into a step: hangs the step, with the steps under it, under another step or branch (for example "impression" instead of "sent"), or changes the link, key or code that a branch listens for.                                                                                                                                                                                                                       |
| `flow_copy_node`                                    | Copies or moves a step, with all the steps under it, under another step.                                                                                                                                                                                                                                                                                                                                                                       |
| `flow_layout`                                       | Places every step of the journey tidily on the canvas in one call, with nothing overlapping. Only positions change.                                                                                                                                                                                                                                                                                                                            |
| `flow_annotation`                                   | Adds, changes, deletes or lists notes on the journey canvas.                                                                                                                                                                                                                                                                                                                                                                                   |
| `set_stop_convert_rule`, `remove_stop_convert_rule` | Create, change or remove the rules that stop a journey or mark it converted when an event happens.                                                                                                                                                                                                                                                                                                                                             |
| `set_campaign_tag`, `remove_campaign_tag`           | Create, change or remove a campaign tag value, used as `{{campaign.<key>}}` in the campaign's messages.                                                                                                                                                                                                                                                                                                                                        |

The assistant usually works in this order: create the message templates (email, push, popup and others) and the segment, create the campaign as a draft, then add the steps, one step or a whole branch per call. The same checks as in the back office apply to every step, and every change is checked on the journey it would leave: a refused change writes nothing. Send the changes of one campaign's journey one at a time, because two changes to the same journey at the same moment can refuse each other.

## Check a change first

* `entity_create` and `entity_update` take an optional `dry_run: true`: the server runs the same checks as the save and answers "ok" (with the change in words and any warnings) or what is wrong. Nothing is saved. Without `dry_run`, the call saves; an update then answers the fields it changed, before and after.
* The `exec_operation` operations take `dry_run: true` too.
* The tools that change data are not marked read-only, so the AI client asks you to confirm each one.

Ask for a confirmation in your request: *"Tell me what you will change, and change nothing until I confirm."*

## Clone a campaign

`clone_campaign` copies a campaign on the same label into a new **Draft**: the journey, the stop and convert rules, and the campaign tags. You can give the copy a name, choose its entry mode, and turn a scheduled campaign into a one-time or a recurring one; the copy keeps the source's user limit per run and its speed. A copy is realtime or scheduled, as its source is. Archived and executed campaigns cannot be edited: clone them to start again.

*"Clone campaign 'Weekend reload' as 'Weekend reload - November', and show me the journey of the copy."*

## Copy a campaign to another label or brand

When each brand has its own label, you can rebuild a campaign from one label on another in one conversation. The connection must cover both labels: choose both on the sign-in page, or create a token for both. The assistant then works on each label with its `label_id`.

`clone_campaign` works inside one label only. A copy to another label is a rebuild, because the ids of segments, templates, bonus templates and events differ between labels:

1. The assistant reads the campaign and its journey on the source label.
2. On the target label, it finds the matching segment, message templates and bonus templates, or creates them.
3. It creates the campaign as a draft on the target label.
4. It adds the journey step by step, and shows you the result as a diagram.
5. You check the draft in the back office of the target label, and make it Active when you are ready.

*"Copy the campaign 'Welcome series' from label A to label B. Use the segments and templates that already exist on label B where they match, create the missing ones, and tell me what you will create before you save it."*

## Start, pause, stop and change the speed

| What                      | How                                                             | Notes                                                                                                                                                                                                                                                                                                    |
| ------------------------- | --------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Start**                 | `entity_update`: set the status to **Active**                   | The AI client asks you to confirm.                                                                                                                                                                                                                                                                       |
| **Pause** or disable      | `entity_update`: set the status to **Paused** or **Disabled**   |                                                                                                                                                                                                                                                                                                          |
| **Stop for every player** | `exec_operation` → `stop_campaign_for_all` with `confirm: true` | The back office's full stop. It stops the journey for every player who is in it now, and cancels their scheduled messages and pending popups and inbox messages. It does not change the status: pause or archive the campaign for that. Without `confirm: true`, the answer only says what would happen. |
| **Change the speed**      | `exec_operation` → `change_campaign_speed`                      | Scheduled campaigns only: players per minute, within the range your label allows.                                                                                                                                                                                                                        |
| **Test with a player**    | `exec_operation` → `user_start_campaign`                        | A test tool: it puts given players into an Active campaign now and skips its trigger, segment and entry conditions. See [MCP - operations on the users](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-operations-on-the-users.md).                                                    |

## Example prompts

* *"Read campaign 'Welcome series' and explain its journey in plain words."*
* *"Create a draft realtime campaign that starts on the first deposit: send the 'Welcome' email, wait 2 days, then send a push to players who did not open it. Tell me the plan first."*
* *"In campaign 'Welcome series', add a 24-hour delay before the push step. Check it with a dry run first."*
* *"Clone campaign 'Weekend reload' and change the copy to run once next Friday at 10:00."*
* *"Copy the journey of 'VIP birthday' from label A to label B, and list the steps for me to confirm first."*
* *"Make the draft campaign 'Weekend reload - November' Active."*
* *"Lower the speed of campaign 'Reactivation' to 500 players per minute."*

## Related pages

* [MCP - Segments management](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-segments-management.md) — build the audience first.
* [MCP - Tools and entity types](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-tools-and-entity-types.md) — every tool and operation.
* [MCP - Golden queries](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-golden-queries) — campaign results, message statistics and players who entered.
* [MCP - Access Profiles](https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-access-profiles.md) — which profiles may create, update and run operations.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.smartico.ai/welcome/technical-guides/mcp-server/mcp-build-and-copy-campaigns-and-journeys.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
