> For the complete documentation index, see [llms.txt](https://docs.pletor.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pletor.ai/automate/pletor-mcp.md).

# Pletor MCP

Build, run, and orchestrate creative workflows from Claude and any MCP-compatible agent.

Connect Pletor to your favorite AI assistant in one click.

Once connected, you can build and edit flows, run them, create assets, and diagnose failures, all from a natural-language conversation.

{% embed url="<https://www.youtube.com/watch?v=PyftxQw0-F8>" %}

## What you can do

<details>

<summary>Create images and videos</summary>

Create images and videos straight from your assistant, no flow required. Describe what you want and Pletor routes it to the right model, refining the brief with you before spending credits. Edit, upscale, or cut out backgrounds on any result. *"make a studio packshot of this product on marble"* comes back as a finished asset, ready to drop into a Brain or a flow.

</details>

<details>

<summary>Build, use and grow your <a href="/build-your-system/brain.md">Brains</a></summary>

Give your assistant your brand's memory. It reads and searches a Brain's products, characters, collections, and Memories, so every request is grounded in your real catalog, identity, and voice. *"use the hero product and our summer palette"* pulls the actual assets and guidelines before creating anything.

It also builds and maintains Brains: spin one up, point it at a website or documents to read, write Memories, products, and characters, and set a brand kit. Everything you encode compounds into every future run.

</details>

<details>

<summary>Build and iterate <a href="/build-your-system/agents.md">Flows</a></summary>

Describe the flow you need and your assistant builds it in Pletor: it creates the flow, adds and wires the nodes, sets models and prompts, and labels each step. You get a Studio link and watch the flow take shape live.

* **Ground it in your Brain.** Bind input nodes to Brain pages so every run pulls your real products and guidelines.
* **Test step by step.** Run one node or a chain of nodes before committing. Results land on the canvas.
* **Edit existing flows.** Swap a model, rewrite a prompt, add a step. Your assistant reserves the flow while it works and hands it back when done.
* **Save versions.** When you're happy, it saves a named version and makes it active, so runs, apps, and the API use it by default.
* **Add custom nodes.** Save a Python script as a reusable node in your library.

*"Build a flow that turns a product photo into three lifestyle shots using our Brain"* comes back as a tested flow in your Studio, ready to run.

</details>

<details>

<summary>Run any <a href="/build-your-system/agents.md">Flow</a> or <a href="/automate/apps.md">App</a></summary>

Trigger a flow or app asynchronously and get a run ID back immediately. Your assistant maps natural-language requests onto the flow's inputs. Runs continue in the background while you keep chatting, and you can fan a flow out across many inputs in one batch.

</details>

<details>

<summary>Find Flows and Apps</summary>

Search your flow and app library by name, tag, flow type, or visibility. Ask *"find a workflow for product photos"* and your assistant returns the right match with its ID, description, and tags, ready to run.

</details>

<details>

<summary>Browse models and pricing</summary>

List every available model with its capabilities, generation time, and credit cost.

</details>

<details>

<summary>Diagnose failed runs in one call</summary>

Ask *"why did my run fail?"* and `diagnose_run` returns a single flat report: the verdict, the failed node, the last successful node, an inputs summary, and a suggested next action.

</details>

<details>

<summary>Search the docs</summary>

Get up-to-date guidance straight from `docs.pletor.ai`. Ask *"how do I structure inputs for the image-to-video node?"* and your assistant pulls the relevant excerpt and answers in context.

</details>

***

## Connect in 3 steps

Pletor MCP works with:

* **Claude** (Desktop, Web, Code)
* **ChatGPT / Codex**
* **Cursor**, **VS Code**, and any other MCP-compatible client

{% hint style="info" %}
**Before you start:** You'll need a paid Pletor account to use the Pletor MCP. See .

It takes a minute, and you'll connect your AI assistant to it in the next steps.
{% endhint %}

{% stepper %}
{% step %}

### Open Connector settings

* **Claude** → [Customize](https://claude.ai/customize/connectors) → Connectors → *Add custom connector*

<figure><img src="https://2795436344-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fnyc5C4KVCklpMy0pLa15%2Fuploads%2Fc3jyQhJ6EdrKuoFGDxTC%2F2026-05-20%2013.57.26.gif?alt=media&#x26;token=ade42580-9b61-458e-80d4-c90a7df2b313" alt=""><figcaption></figcaption></figure>

* **ChatGPT** → [Plugins](https://chatgpt.com/plugins) → *Add MCP server*
* **Cursor / VS Code** → MCP Servers → *Add server*
  {% endstep %}

{% step %}

### Add the Pletor MCP URL

Name the connector **Pletor** and paste this URL:

```
https://api.pletor.ai/mcp
```

{% endstep %}

{% step %}

### Sign in with your Pletor account

Click **Add** → **Connect**. You'll be redirected to log in with your existing Pletor account. Once authorized, the connector is ready.
{% endstep %}
{% endstepper %}

{% hint style="info" %}
**Can't find "custom connectors"?**\
\
It's likely your organization does not currently allow for new connectors. Contact your workspace admin.
{% endhint %}

{% hint style="info" %}
**No API keys to manage.** Pletor MCP uses OAuth: your assistant signs in on your behalf and inherits your organization's permissions and credit balance.
{% endhint %}

***

## From idea to output in one conversation

### Creative ops & automation

*"Build a flow that takes a product photo, writes the prompt from our Brain, and renders four ad variants. Test it on one product, then save it."*\
Your assistant builds the flow live in your Studio, tests each step, and saves a version your team can run from Pletor, an app, or the API.

### Marketing & content teams

*"Run my UGC ad workflow on this product photo and give me three variants."*\
Pletor uploads the asset, kicks off the workflow, and surfaces the creatives, all in chat.

### E-commerce

*"Create lifestyle shots for product A using our Acme Brain."*\
Your assistant searches the Acme Brain, retrieves style guidelines and product A references, runs the shots with the right inputs and model, and returns the results without leaving the chat.

### Agencies & freelancers

*"Show me every video workflow my team has built, then run the storyboard one on this brief."*\
Browse, pick, and execute across your whole library in seconds.

### Filmmakers & creators

*"Take this character reference and run the cinematic shot workflow with a sunset lighting prompt."*\
Upload, configure, and render without touching the Pletor UI.

***

## Example prompts

### Build

```
Build a flow that turns a product photo into 3 lifestyle shots using my Acme Brain
```

```
Add an upscale step after the image node in my "Lifestyle shot" flow, then test it
```

```
Swap the video model in my UGC ad flow to Kling 3 and save it as a new version
```

### Discover

```
Find my workflows tagged "video"
```

```
List public workflows for the standard flow type
```

```
What image models do you support, and which are cheapest?
```

### Run

```
Run workflow abc-123 with prompt "a cat on a roof at golden hour"
```

```
Upload https://example.com/product.png and run my "lifestyle shot" workflow with it
```

```
Create 3 variants of my UGC ad using this brief: [...]
```

***

## FAQ

<details>

<summary>How does Pletor connect to my AI assistant?</summary>

Pletor exposes an MCP (Model Context Protocol) server at `https://api.pletor.ai/mcp`. Any MCP-compatible client (Claude, Codex, Cursor, VS Code) can connect to it as a remote connector. Auth is handled via OAuth, so you sign in with your Pletor account and never copy-paste keys.

</details>

<details>

<summary>Which assistants are supported?</summary>

Any client that supports remote MCP servers: Claude (Desktop, Web, Code), Codex with MCPs, Cursor, VS Code, and others as the ecosystem grows.

</details>

<details>

<summary>Do I need an API key?</summary>

No. Pletor MCP uses OAuth: log in once with your Pletor account and you're connected. Your assistant inherits your organization's workspace, workflows, and credit balance.

</details>

<details>

<summary>How does billing work?</summary>

Runs and uploads consume credits from your existing Pletor plan, just like the web app. Building and editing flows, browsing workflows, inspecting definitions, and searching docs are free. Testing nodes while building consumes credits like any run. There is no separate MCP subscription.

</details>

<details>

<summary>How long does a run take?</summary>

Runs are asynchronous. `run_workflow` returns a `flow_run_id` immediately; your assistant polls `diagnose_run` or `get_flow_run_status` until the run reaches `completed`, `failed`, or `canceled`. Duration depends on the workflow: single image nodes finish in seconds, multi-step video workflows can take minutes.

</details>

<details>

<summary>Can I run workflows I built in Pletor?</summary>

Yes. Every workflow in your organization (private, shared, or public) is visible through the MCP. The MCP is a remote control for your existing Pletor workspace, not a separate environment.

</details>

<details>

<summary>Can my assistant edit a flow I'm working on?</summary>

Yes. Before editing an existing flow, your assistant reserves it so its changes and yours don't collide, then releases it when done. Every edit shows up live in Pletor Studio, and nothing becomes the active version until it's saved.

</details>

<details>

<summary>Can I upload files from my computer?</summary>

Yes. In Claude and ChatGPT, your assistant opens a drop zone right in the chat. In CLI agents like Claude Code or Codex, it uploads the file for you. Files already hosted on a public URL can be passed directly.

</details>

<details>

<summary>Is it secure?</summary>

Yes. OAuth tokens are scoped to your user and validated against Pletor's auth provider on every request. Tokens never leave your assistant; Pletor only receives the Bearer token the client sends. Asset visibility (`private` / `shared` / `public`) follows the same rules as in the Pletor web app.

</details>