Zyte MCP

npx add-mcp 'https://mcp.zyte.com/v1/mcp'

Zyte MCP

Zyte MCP is a remote MCP server that lets an agent use Zyte on your behalf: to get web pages, to extract data from them, to search the web, and to work with your Scrapy Cloud projects. to get started.

Some Zyte MCP capabilities require a Zyte API key. New users can sign up on the Zyte dashboard or when they connect Zyte MCP, and receive $5 in Zyte API trial credit for one month.

Connect your agent

Zyte MCP is a remote MCP server at https://mcp.zyte.com/v1/mcp, which you can use from any MCP client. To connect it, follow the installation instructions for your client below:

Run the following in a terminal:

claude mcp add --transport http zyte https://mcp.zyte.com/v1/mcp

Then sign in to Zyte:

claude mcp login zyte

Run the following in a terminal:

codex mcp add zyte --url https://mcp.zyte.com/v1/mcp

Then sign in to Zyte:

codex mcp login zyte

Run the following in a terminal:

copilot mcp add --transport http zyte https://mcp.zyte.com/v1/mcp

Then sign in to Zyte:

copilot mcp login zyte

Run the following in a terminal:

code --add-mcp '{"name":"zyte","type":"http","url":"https://mcp.zyte.com/v1/mcp"}'

Then, in VS Code, start the server from the MCP Servers view, and sign in to Zyte when prompted.

Add Zyte MCP as a custom connector:

1. Open Settings, then Connectors.
2. Select Add custom connector.
3. Set the name to Zyte and the URL to https://mcp.zyte.com/v1/mcp, and add it.
4. Select Connect, and sign in to Zyte.

Most MCP clients are configured through a JSON file with a section like the following:

{
  "mcpServers": {
    "zyte": {
      "type": "http",
      "url": "https://mcp.zyte.com/v1/mcp"
    }
  }
}

Then follow your client’s instructions to sign in to the server.

Authentication

To use Zyte MCP with your Zyte account, you need to sign in. Your client’s sign-in process may differ; see the installation instructions above for your client.

When you sign in, your agent makes Zyte API requests with a dedicated API key of your organization, mcp_access_key, which you can use to. If you have access to more than one organization, your agent asks you which one to use, see.

Check it works

Once connected, ask your agent to use Zyte for something small, naming Zyte so that it does not use its own tools:

Fetch https://example.com with Zyte and tell me the title of the page.

Your agent should answer Example Domain. To confirm that Zyte MCP did the work, and not your agent’s own web access:

• Look at the tool call in your client. It should show a call to a Zyte MCP tool, fetch_http in this case. Many clients ask you to approve the call the first time.
• Check your Zyte API stats. The request shows up shortly after, and you can filter the stats by the mcp_access_key API key.

If your agent answers without calling a Zyte tool, name Zyte more explicitly, see. If it cannot connect, see.

What you can ask

Ask for the outcome, and let your agent choose the tools. For example:

• “Fetch https://example.com/some-page with Zyte.”
• “Open this page in a browser with Zyte, accept the cookie banner and take a screenshot.”
• “Extract the name, price and availability of the product at <URL>.”
• “Search the web for ‘zyte api pricing’ as seen from Germany.”
• “Does Zyte API support example.com, and how much does a request to it cost?”
• “Why did the last job of my ``books`` spider fail?”
• “Run my ``books`` spider every Monday at 09:00 UTC.”
• “How much did I spend on Zyte API this week, by domain?”

Working with your agent

Agents have their own tools to get web pages and search the web, and often use those instead of Zyte. Naming Zyte in your request, e.g. “fetch <URL> with Zyte”, is usually enough to get Zyte used. For pages that need a browser or a screenshot, or for data extraction, agents tend to choose Zyte on their own.

Your agent drives the whole task, and Zyte handles one request at a time. How many requests a task takes, and what your agent does with what comes back, is up to your agent.

[!warning] Warning Your agent can act on the Scrapy Cloud projects that your account can access. Among other things, it can start and cancel jobs, create, change and delete schedules, change project settings, and write and delete collection records. Writing a record replaces any record with the same key. Deleting a schedule, a collection or its records cannot be undone. Review what your agent asks to do before you approve it.

Tools

Zyte MCP gives your agent the following tools. You do not need to name them: your agent picks the right one for your request.

Zyte API and account tools:

Tool	What it does	Learn more
fetch_http	Get a URL with a plain HTTP request, avoiding bans, with any method, headers and body: APIs and forms as well as static pages.	HTTP requests, bans
fetch_page	Get a URL in a browser, as rendered HTML, a screenshot or both. Your agent can interact with the page first: click, type, scroll, wait for an element, run JavaScript, and so on. It can, for example, log into a website and continue from there.	Browser requests, screenshots, actions
extract_from_http	Extract structured data, with AI, from a page downloaded with a plain HTTP request: the fields of a product, an article, a job posting or other page types, the items on a listing page, or the links to follow to keep crawling a website. Your agent can also describe the fields it wants in plain words, for data that no standard page type covers.	Automatic extraction, page types
extract_from_browser	The same, from a page rendered in a browser, optionally after browser actions.	Automatic extraction
search	Search the web and get structured results, optionally as seen from a given country or language.	Search, location
zyte_api_domain_pricing	Check whether Zyte API supports a website and what a request to it costs.	Pricing
zyte_api_usage_stats	Report your recorded Zyte API usage: requests, cost, response times and status codes over a time window, as a summary or broken down by date, domain or feature.	Stats API, features
user_info	List the Zyte organizations you belong to.

The fetch and extract tools also let your agent choose the location and the type of IP address of the request, send and receive cookies, and keep a session across requests. The fetch tools can also tag requests so that you can find them later in your stats, and the plain HTTP tools can emulate a mobile device.

Scrapy Cloud tools:

Tool	What it does	Learn more
scrapy_cloud_list_projects, scrapy_cloud_get_project, scrapy_cloud_get_project_activity	Find projects and read their details and activity log.	Projects
scrapy_cloud_list_spiders, scrapy_cloud_run_spider, scrapy_cloud_run_script	List spiders, and start a spider or a script.	Spiders, scripts
scrapy_cloud_list_jobs, scrapy_cloud_get_job, scrapy_cloud_cancel_job, scrapy_cloud_update_job_tags	Find jobs, read their state and counters, cancel them and tag them.	Jobs
scrapy_cloud_get_job_log, scrapy_cloud_get_job_requests, scrapy_cloud_get_job_items, scrapy_cloud_get_job_item_stats	Read the log, requests and items of a job, and how many items have each field.	Logs, items
scrapy_cloud_list_periodic_jobs, scrapy_cloud_get_periodic_job, scrapy_cloud_create_periodic_job, scrapy_cloud_update_periodic_job, scrapy_cloud_delete_periodic_job	Read, create, pause, change and delete scheduled jobs.	Periodic jobs
scrapy_cloud_get_project_settings, scrapy_cloud_update_project_settings	Read and change project settings.	Projects
scrapy_cloud_list_collections, scrapy_cloud_get_collection_records, scrapy_cloud_put_collection_record, scrapy_cloud_delete_collection_records, scrapy_cloud_delete_collection	Read, write and delete collections and their records.	Collections

How it works

1. You and, the first time it uses Zyte MCP,.
2. When you ask for something, your agent picks a and calls it, one request at a time, and works with the result.
3. Everything your agent does is billed to your organization. See.

Billing

Using Zyte MCP is free, but what your agent does through it is billed to your Zyte account at Zyte API prices and counts towards your rate limits. A single request to your agent can result in many Zyte API requests, for example one per page of a paginated listing. Jobs that your agent starts on Scrapy Cloud are billed at Scrapy Cloud prices.

Your agent’s Zyte API requests use the mcp_access_key API key. Filter your Zyte API stats by that key to see what your agent spent, and set a spending limit on it to cap that spending. Jobs that your agent starts on Scrapy Cloud are not covered by either.

Organizations

If your Zyte account gives you access to more than one organization, your agent asks you which one to use, and everything it does through Zyte MCP is billed to that organization. Your agent keeps using it for the rest of the session, unless you ask for a different one.

To avoid being asked in every session, name the organization in the instructions that your agent reads at startup, e.g. in AGENTS.md or CLAUDE.md:

Use the Zyte organization with ID 123456.

Put that line in the instructions of each project to use a different organization per project, or in your global instructions to use the same one everywhere. Agents with memory, like Claude Code, may also remember it if you tell them, e.g. “always use Zyte organization 123456”.

To find the ID of an organization, ask your agent to list your Zyte organizations, or open the organization on the Zyte dashboard: its ID is the number in the URL, https://app.zyte.com/o/<ID>.

Troubleshooting

My agent does not use Zyte

See.

My organization is not offered

Only organizations with an active Zyte API or Scrapy Cloud plan are listed. Check the plans of the organization on the Zyte dashboard.

My agent asks which organization to use in every session

See to name it in your agent instructions.

My agent used more Zyte API requests than I expected

See.

My client reports 401 or “Invalid Zyte API credential”

Your sign-in was not accepted. Sign in again. See.

A tool reports “Not found” or an invalid argument

The project, job, collection or schedule that your agent asked for does not exist, or your account cannot access it. Ask your agent to list your projects, spiders or collections, and to try again with an ID from the list.

My agent is rate limited

Zyte MCP requests count towards your rate limits. Zyte API may respond with a 429 or 503 status code, and you are not charged for those responses. Wait and try again, or ask your agent to make fewer requests at once. To raise your limits, see Increasing rate limits. Usage stats requests are limited to 20 per minute.

A request is slow or times out

Browser requests, and search, take seconds, while plain HTTP requests usually take less than a second. If a page does not need a browser, ask your agent for a plain HTTP request. See server timeouts for what to expect from Zyte API.

I need help from support

Open a support ticket and include the tool your agent used, the error message, the time in UTC, your organization ID, and the request_id of the result, if it has one.