---
title: "ChatGPT · Scrape.ly Docs"
description: "Add Scrape.ly to ChatGPT as a custom MCP server."
url: https://scrape.ly/docs/chatgpt
markdown: https://scrape.ly/docs/chatgpt.md
---

# ChatGPT

> Add Scrape.ly to ChatGPT as a custom MCP server.

Add Scrape.ly to ChatGPT as a custom MCP server: in a conversation, ChatGPT can search the Store, run Actors, wait for their runs, and read the results. It connects to the hosted [MCP server](https://scrape.ly/docs/mcp.md) and signs in with OAuth, so there is no token to copy.

Which accounts can add custom MCP servers depends on your plan and, in a Business or Enterprise workspace, on what your admin allows. If you don't see the option below, check OpenAI's help article on developer mode or ask your workspace admin.

## Add the server

1. Open [ChatGPT Plugins](https://chatgpt.com/plugins), select the plus button, then **Add custom MCP server**.
2. Name it `Scrape.ly` and add a short description, such as `Run Scrape.ly Actors and read their results`.
3. Under **Connection**, choose **Public endpoint** and enter `https://mcp.scrape.ly`.
4. Set authentication to **OAuth**, review the warning, and select **I understand and want to continue**.
5. Select **Create as a plugin**. A dashboard window opens where you approve access, then ChatGPT lists the server's tools.
6. Install the plugin.

ChatGPT can do anything your account can, including starting paid runs. A **Max charge per day** applies only to [API keys](https://scrape.ly/docs/api-keys.md), not to an OAuth sign-in, so put limits in your request.

## Use it in a conversation

Start a new conversation, type `@`, pick **Scrape.ly**, and ask:

```text
Find an Actor that scrapes Google Maps, run it for 'coffee shops in Berlin' with at most 20 results, and list the name, rating, and address of each place.
```

## What the assistant can do

The assistant gets the MCP server's default tools. The ones used most:

| Tool | What it does |
| --- | --- |
| `search-actors` | Find Actors in the Store by keywords. |
| `fetch-actor-details` | Read an Actor's input schema, README, and pricing before running it. |
| `call-actor` | Run an Actor with a JSON input, optionally with `maxItems` and `maxTotalChargeUsd`. |
| `get-actor-run` | Wait for a run and check its status. |
| `get-actor-output` | Read a run's results, paged and kept compact. |

The full list, including saved inputs and automations, is on the [MCP server](https://scrape.ly/docs/mcp.md#tools) page. To run one Actor you already know, name it in the request, for example `Run username/actor-name`, so the assistant skips the search.

**Costs**: runs started by the assistant are billed like any run. Put the limits in the request (such as "at most 20 results"): an OAuth sign-in has no daily spending cap.

---

- Previous: [Amazon Bedrock](https://scrape.ly/docs/amazon-bedrock.md)
- Next: [Claude Desktop](https://scrape.ly/docs/claude-desktop.md)
- All docs: [Docs index](https://scrape.ly/docs.md), [llms.txt](https://scrape.ly/llms.txt)