MCP tool reference - Zenrows Docs

This page lists the tools the Zenrows MCP server exposes at https://mcp.zenrows.com/mcp, how each one is classified, and the parameters, outputs, and errors of the scrape, extract, and account_usage tools. For setup, see Remote MCP server.

Tool classification

The classification comes from each tool’s MCP annotations. Read-only tools retrieve information and change nothing. Write tools change state: a job on your account, or a page inside a cloud browser session. Only batch_cancel is annotated as destructive.

Tool	Classification	Acts on
`scrape`	Read-only	Reads one web page
`extract`	Read-only	Reads one web page
`account_usage`	Read-only	Reads your plan and credit usage
`batch_create`	Write	Creates a Batch job on your account
`batch_status`	Read-only	Reads a Batch job
`batch_results`	Read-only	Reads a Batch job’s results
`batch_wait`	Read-only	Reads a Batch job until it finishes
`batch_cancel`	Write, destructive	Stops a running Batch job
`browser_navigate`	Write	Opens a cloud browser session or navigates it
`browser_close`	Write	Closes a cloud browser session
`browser_go_back`	Write	Navigates a browser session back
`browser_go_forward`	Write	Navigates a browser session forward
`browser_reload`	Write	Reloads the page in a browser session
`browser_click`	Write	Clicks an element on the page
`browser_hover`	Write	Hovers over an element on the page
`browser_type`	Write	Types text into an element
`browser_fill`	Write	Sets the value of an input
`browser_select_option`	Write	Selects an option in a dropdown
`browser_check`	Write	Checks a checkbox
`browser_uncheck`	Write	Unchecks a checkbox
`browser_focus`	Write	Focuses an element
`browser_press_key`	Write	Presses a key on the page
`browser_scroll`	Write	Scrolls the page
`browser_drag`	Write	Drags an element
`browser_evaluate`	Write	Runs JavaScript on the page
`browser_set_cookies`	Write	Sets cookies in a browser session
`browser_clear_cookies`	Write	Clears cookies in a browser session
`browser_local_storage`	Write	Reads or writes local storage in a browser session
`browser_new_tab`	Write	Opens a tab in a browser session
`browser_switch_tab`	Write	Switches tabs in a browser session
`browser_batch`	Write	Runs several browser actions in one call
`browser_get_accessibility_tree`	Read-only	Reads the page’s accessibility tree
`browser_get_url`	Read-only	Reads the current URL
`browser_get_title`	Read-only	Reads the page title
`browser_get_text`	Read-only	Reads text content
`browser_get_attribute`	Read-only	Reads an element attribute
`browser_get_html`	Read-only	Reads HTML
`browser_query_selector_all`	Read-only	Reads all elements matching a selector
`browser_screenshot`	Read-only	Captures a screenshot
`browser_generate_pdf`	Read-only	Renders the page as a PDF
`browser_wait_for_selector`	Read-only	Waits for an element
`browser_wait_for_navigation`	Read-only	Waits for a navigation to finish
`browser_wait`	Read-only	Waits a fixed time
`browser_get_cookies`	Read-only	Reads cookies

Descriptions of the Batch and browser tools are in Batch tools and Browser tools.

Authentication and permissions

The server accepts either of these on every request:

OAuth 2.1 with PKCE (S256). Clients discover the authorization server through https://mcp.zenrows.com/.well-known/oauth-protected-resource and can register dynamically at https://mcp.zenrows.com/register. The user signs in at app.zenrows.com, or creates a free account, and approves access.
API key as a Bearer token, in the Authorization header. See Authentication.

There is one scope, api. The access token is the account’s API key, so every tool runs with the same permissions as that key. To revoke access, rotate the API key in the dashboard: calls with the old key then fail with AUTH003.

Credits and limits

The Free plan includes 5,000 credits that renew each month; paid plans include larger monthly allowances. Each request costs credits depending on the site and the options used. See Pricing for how each tool is billed and Plans and pricing for plan allowances. When the credits run out, requests fail with AUTH004 until the allowance renews, or the account tops up or upgrades. A key can also have its own credit cap, which returns AUTH014. Each plan also limits how many requests run at once. Requests over that limit fail with AUTH006. account_usage does not count toward the concurrency limit.

Errors

A failed tool call returns a tool result with isError: true. The text of the result contains the Zenrows error as JSON, with a code, a title, and a detail that explains the fix.

Code	Meaning	Retry?
`AUTH003`	The API key is not valid, for example after it was rotated	No. Reconnect with a valid key
`AUTH004`	The account’s credit allowance is spent	No. Wait for the renewal or upgrade
`AUTH006`	Too many requests running at once	Yes, after current requests finish
`AUTH014`	This API key reached its credit cap	No. Raise the cap or wait for its window
`REQS007`	Extract has not prepared this domain yet (`extract` in `auto` mode)	No. `extract` already retries once with `autoparse`; you only see this code when `fallback_autoparse` is `false`

The full list is in API error codes.

`scrape`

Fetches one web page and returns its content as Markdown, plain text, HTML, a PDF, a screenshot, or JSON. Built on Fetch. Requests use Adaptive Stealth Mode (mode=auto) by default: Zenrows enables JavaScript rendering and premium proxies only when a page needs them, and you pay only for the configuration that succeeds. Setting js_render or premium_proxy forces that configuration and turns Adaptive Stealth Mode off. Classification: read-only. Makes no changes to the account or to the site it reads.

Parameters

string

required

The URL of the page to fetch.

boolean

default:"false"

Force rendering in a headless browser on every request. Turns Adaptive Stealth Mode off, which already renders when a page needs it.

boolean

default:"false"

Force residential proxies on every request. Costs more credits and turns Adaptive Stealth Mode off, which already switches to them when a site blocks.

string

Two-letter country code (ISO 3166-1 alpha-2), for example US. Works with Adaptive Stealth Mode. Without it (when js_render is set), it requires premium_proxy.

string

default:"markdown"

One of markdown, plaintext, pdf, or html. Ignored when autoparse, css_extractor, outputs, or a screenshot option is set.

boolean

Return the page’s main data as JSON.

string

JSON object mapping field names to CSS selectors, for example {"title":"h1","price":".price"}. Returns JSON.

string

Comma-separated data types to return as JSON: emails, headings, links, menus, images, videos, audios, or * for all.

string

CSS selector to wait for before capturing. Requires js_render; Adaptive Stealth Mode alone ignores it.

integer

Milliseconds to wait after the page loads, up to 30000. Requires js_render; Adaptive Stealth Mode alone ignores it.

string

JSON array of browser actions to run before capturing, for example [{"click":"#load-more"},{"wait":1000}]. Works with Adaptive Stealth Mode, which then renders the page in a browser, or with js_render.

boolean

Return a screenshot of the visible part of the page instead of text.

boolean

Return a screenshot of the full page.

string

Return a screenshot of the element matching this CSS selector.

Output

The page content in the requested format: text for markdown, plaintext, and html; JSON for autoparse, css_extractor, and outputs; an image for screenshots.

Example

Call

{ "name": "scrape", "arguments": { "url": "https://example.com" } }

Result

This domain is for use in documentation examples without needing permission. This is not a service; avoid relying on it for testing and monitoring purposes.

`extract`

Returns a page as structured JSON fields instead of a full page body, for example a product’s name and price. Built on Extract. In its default auto mode, extract works on domains Extract has prepared. The list is available from GET /v1/extract/domains. On other domains it retries once with autoparse. Like scrape, every mode uses Adaptive Stealth Mode (mode=auto) by default. Setting js_render or premium_proxy forces that configuration and turns it off. Classification: read-only. Makes no changes to the account or to the site it reads.

Parameters

string

required

The URL of the page to extract from.

string

default:"auto"

auto uses site-tailored extraction (extract=auto, open beta) on prepared domains. autoparse (deprecated) returns general-purpose JSON on any domain. css uses your own selectors from css_extractor.

string

Required when mode is css. JSON object mapping field names to CSS selectors.

boolean

Force rendering in a headless browser on every request. Turns Adaptive Stealth Mode off.

boolean

Force residential proxies on every request. Costs more credits and turns Adaptive Stealth Mode off.

string

Two-letter country code. Works with Adaptive Stealth Mode. Without it (when js_render is set, or mode_auto is false), it requires premium_proxy.

boolean

default:"true"

Adaptive Stealth Mode (mode=auto) is on by default. Set false to send a plain request without it.

string

CSS selector to wait for before extracting. Requires js_render; Adaptive Stealth Mode alone ignores it.

integer

Milliseconds to wait after the page loads, up to 30000. Requires js_render; Adaptive Stealth Mode alone ignores it.

boolean

default:"true"

When mode is auto and the domain isn’t enabled (AUTH010) or prepared (REQS007) for Extract, retry once with autoparse. See Errors.

Output

A JSON object:

Field	Type	Description
`ok`	boolean	`true` when the extraction ran
`mode`	string	The mode that produced the data
`fellBackToAutoparse`	boolean	`true` when `auto` fell back to `autoparse`
`empty`	boolean	`true` when no field came back with a value, for example `{ "listings": [] }`
`data`	object or array	The extracted fields, or `null`

Example

Call

{ "name": "extract", "arguments": { "url": "https://www.amazon.com/s?k=airpods+pro" } }

Result (trimmed)

{
  "ok": true,
  "mode": "auto",
  "fellBackToAutoparse": false,
  "empty": false,
  "data": {
    "pagination": { "current_page": 1, "total_pages": 20 },
    "results": [
      { "position": 1, "asin": "B0FQFB8FMG", "price": 179, "list_price": 249 }
    ]
  }
}

`account_usage`

Returns the current plan, its credit allowance, how much is spent, and when the period renews. Free, and does not count toward the concurrency limit. Classification: read-only. Makes no changes to the account.

Parameters

None.

Output

The account’s subscription details as JSON, passed through from the Zenrows API. Fields include status, period_starts_at, period_ends_at, usage_credits, credit_limit, usage_percent, plan (name, products, concurrency limit), and api_key.caps.

Example

Call

{ "name": "account_usage", "arguments": {} }

Result (trimmed)

{
  "status": "TRIALING",
  "period_ends_at": "2026-11-05T14:25:43Z",
  "usage_credits": 0,
  "credit_limit": 5000,
  "usage_percent": 0,
  "plan": {
    "name": "Free",
    "products": { "api": { "concurrency": { "limit": 5, "usage": 0 } } }
  },
  "api_key": { "caps": [] }
}

​Tool classification

​Authentication and permissions

​Credits and limits

​Errors

​scrape

​Parameters

​Output

​Example

​extract

​Parameters

​Output

​Example

​account_usage

​Parameters

​Output

​Example

Tool classification

Authentication and permissions

Credits and limits

Errors

`scrape`

Parameters

Output

Example

`extract`

Parameters

Output

Example

`account_usage`

Parameters

Output

Example