MCP Tools Reference

Complete reference for all 33 tools available through the Scrayle MCP Server. Each tool is callable by your AI assistant when connected via MCP.

Scraping Tools

`scrape_url`

Scrape a single URL and extract content as markdown, HTML, links, or screenshots.

Parameter	Type	Required	Description
`url`	string (URL)	Yes	The page URL to scrape
`formats`	array	No	Output formats: `markdown`, `html`, `links`, `screenshot`. Default: `["markdown"]`
`waitFor`	number	No	Wait time in ms after page load (0-10000)
`mobile`	boolean	No	Emulate a mobile viewport

`batch_scrape`

Scrape multiple URLs in parallel (up to 20 at once).

Parameter	Type	Required	Description
`urls`	string[] (URLs)	Yes	Array of URLs to scrape (max 20)
`formats`	array	No	Output formats. Default: `["markdown"]`

Crawling

`crawl_website`

Recursively crawl a website starting from a URL, following links and extracting content from each page.

Parameter	Type	Required	Description
`url`	string (URL)	Yes	Starting URL for the crawl
`maxPages`	number (1-500)	No	Maximum pages to crawl. Default: 50
`maxDepth`	number (1-10)	No	Maximum link depth. Default: 3
`allowedDomains`	string[]	No	Only follow links to these domains
`excludePaths`	string[]	No	URL path patterns to skip

Browser Control Tools

`start_browser_session`

Launch a cloud browser session and connect to it. Returns a task ID for use with all other browser_* tools.

Parameter	Type	Required	Description
`browserMode`	enum	No	`private`, `stealth`, or `human`. Default: `stealth`
`proxyCountry`	string	No	ISO country code for proxy (e.g. `US`, `GB`)
`sessionTTL`	number (10-7200)	No	Session time-to-live in seconds. Default: 300
`profileId`	string	No	Browser profile ID for persistent cookies/storage

`browser_navigate`

Navigate the browser to a URL. Returns the page title and URL after navigation.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID from start_browser_session
`url`	string (URL)	Yes	The URL to navigate to
`waitUntil`	enum	No	`load`, `domcontentloaded`, or `networkidle`. Default: `domcontentloaded`

`browser_click`

Click an element on the page by CSS selector.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`selector`	string	Yes	CSS selector of the element to click

`browser_type`

Type text into an input field. Optionally clear the field first and press Enter after typing.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`selector`	string	Yes	CSS selector of the input element
`text`	string	Yes	The text to type
`clear`	boolean	No	Clear the field before typing. Default: true
`pressEnter`	boolean	No	Press Enter after typing. Default: false

`browser_screenshot`

Take a screenshot of the current page. Returns a base64 PNG image.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`fullPage`	boolean	No	Capture the full scrollable page. Default: false (viewport only)

`browser_get_content`

Get the visible text content of the page or a specific element.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`selector`	string	No	CSS selector to extract content from. Default: entire page body

`browser_evaluate`

Execute JavaScript in the browser page context. Returns the result.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`script`	string	Yes	JavaScript code to execute in the page context

`browser_select`

Select an option from a dropdown/select element.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`selector`	string	Yes	CSS selector of the select element
`value`	string	Yes	The option value or visible label to select

`browser_scroll`

Scroll the page up or down.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`direction`	enum	Yes	`up` or `down`
`pixels`	number (100-5000)	No	Pixels to scroll. Default: 500

`browser_wait`

Wait for an element to appear on the page, or for the page to finish loading.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID
`selector`	string	No	CSS selector to wait for (waits until visible)
`timeout`	number (1000-30000)	No	Max wait time in ms. Default: 10000

`close_browser_session`

Close an active cloud browser session and disconnect.

Parameter	Type	Required	Description
`taskId`	string	Yes	The browser session task ID

`list_browser_profiles`

List your saved browser profiles. No parameters required.

Workflow Tools

`create_workflow`

Create a new automation workflow with nodes and edges.

Parameter	Type	Required	Description
`name`	string	Yes	Name for the workflow
`nodes`	array	Yes	Array of workflow nodes. Each node has `id`, `type` (e.g. `Scraper`, `DataExtractor`), and `data` object
`edges`	array	Yes	Array of edges connecting nodes. Each edge has `source` and `target` node IDs
`description`	string	No	Optional description of the workflow

`list_workflows`

List your automation workflows. No parameters required.

`run_workflow`

Execute a visual automation workflow by its ID.

Parameter	Type	Required	Description
`workflowId`	string	Yes	The workflow ID to run
`workflowName`	string	No	A human-readable name for this run
`inputParams`	object	No	Key-value input parameters for the workflow
`proxyCountry`	string	No	ISO country code for proxy
`browserMode`	enum	No	`private`, `stealth`, or `human`

`get_workflow_status`

Check the status of a workflow run.

Parameter	Type	Required	Description
`taskId`	string	Yes	The workflow run task ID

`list_workflow_runs`

List your recent workflow runs with their status.

Parameter	Type	Required	Description
`limit`	number (1-50)	No	Number of runs to return. Default: 10

Marketplace Tools

`search_marketplace`

Search the Scrayle Marketplace for pre-built scraping workflows and automation actors.

Parameter	Type	Required	Description
`query`	string	No	Search query (e.g. "tiktok scraper", "amazon products")
`category`	string	No	Filter by category (e.g. "social-media", "e-commerce")
`listingType`	enum	No	`template` or `actor`
`limit`	number (1-50)	No	Number of results. Default: 20

`get_marketplace_actor`

Get detailed information about a marketplace actor/template.

Parameter	Type	Required	Description
`templateId`	string	Yes	The marketplace template/actor ID

`run_marketplace_actor`

Run a marketplace actor with custom input parameters.

Parameter	Type	Required	Description
`templateId`	string	Yes	The actor/template ID to run
`input`	object	No	Input parameters for the actor

CAPTCHA Tools

`solve_captcha`

Solve a CAPTCHA challenge and return the solution token.

Parameter	Type	Required	Description
`type`	string	Yes	CAPTCHA type (e.g. `ReCaptchaV2TaskProxyless`, `HCaptchaTaskProxyless`, `TurnstileTaskProxyless`)
`websiteURL`	string (URL)	Yes	The page URL with the CAPTCHA
`websiteKey`	string	Yes	The CAPTCHA site key
`isInvisible`	boolean	No	Whether the CAPTCHA is invisible (reCAPTCHA only)

Storage Tools

`kv_create_namespace`

Create a new key-value store namespace.

Parameter	Type	Required	Description
`name`	string	Yes	Name for the new KV namespace

`kv_get`

Get a value from a key-value store namespace.

Parameter	Type	Required	Description
`namespaceId`	string	Yes	The KV namespace ID
`key`	string	Yes	The key to retrieve

`kv_set`

Set a value in a key-value store namespace.

Parameter	Type	Required	Description
`namespaceId`	string	Yes	The KV namespace ID
`key`	string	Yes	The key to set
`value`	any	Yes	The value to store (any JSON-serializable value)
`ttl`	number	No	Time-to-live in seconds

`kv_list_namespaces`

List your KV store namespaces. No parameters required.

`create_dataset`

Create a new dataset for storing structured records.

Parameter	Type	Required	Description
`name`	string	Yes	Name for the new dataset

`list_datasets`

List your datasets with item counts. No parameters required.

`get_dataset_items`

Get items from a dataset.

Parameter	Type	Required	Description
`datasetId`	string	Yes	The dataset ID
`limit`	number (1-100)	No	Number of items. Default: 25
`offset`	number	No	Pagination offset. Default: 0

`add_dataset_items`

Add items to a dataset.

Parameter	Type	Required	Description
`datasetId`	string	Yes	The dataset ID
`items`	object[] (1-100)	Yes	Array of data items (flat key-value objects)

💡Tool Usage

You do not call these tools directly. Your AI assistant calls them automatically based on your conversation. Just describe what you need in natural language.