NagentArtefacts
Dev & Technology
Enterprise and White Glove
Company
Scrapingbee connects to a Nagent workspace with an API key. Once it is connected, agents can call 5 Scrapingbee actions, such as "ScrapingBee Data Extraction", "ScrapingBee HTML Fetch" and "ScrapingBee Proxy Mode". Nothing is enabled on connect: each action is allowed one at a time, and an action with side effects runs or waits for a person according to the agent's level.
Every operation an agent can call against Scrapingbee, with input parameters and output schema.
SCRAPINGBEE_DATA_EXTRACTIONTool to extract structured data from a webpage using CSS or XPath selectors. Use ScrapingBee's extract_rules feature.
Input parameters
The webpage URL to extract data from.
Seconds to wait before extraction (for dynamic content).
Emulate device type (desktop or mobile).
Your ScrapingBee API key.
JSON object defining fields to extract and their CSS/XPath selectors. For nested selectors, use object with 'selector' and optional 'type' keys. Misaligned or invalid selectors silently drop fields with no error — verify each selector matches the target DOM before large-scale use.
Whether to render JavaScript before extraction.
Two-letter country code for proxy geolocation (e.g., 'us', 'de').
Use premium proxy for higher reliability.
Block images, CSS, and resources to speed up extraction.
Custom HTTP headers to forward to the target website. Provide as a dict, e.g., {'Accept-Language': 'en-US'}. Headers will be prefixed with 'Spb-' and forwarded to the target.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
SCRAPINGBEE_HTML_FETCHTool to fetch HTML or screenshot via ScrapingBee HTML API. Use when you need page markup or image after optional JS rendering and resource controls. For anti-bot or CAPTCHA-protected sites (e.g., Cloudflare), combine render_js=true with premium_proxy=true or stealth_proxy=true to avoid blocks.
Input parameters
The URL to scrape.
Milliseconds to wait before returning content.
Number of retries on request failure.
Device type to emulate ('desktop' or 'mobile').
Cookies to send in requests (HTTP header string).
CSS selector to wait for before returning content.
Block ads and tracking scripts.
Render JavaScript before returning HTML. Required for client-side rendered pages where dynamic data is absent in raw HTML.
JavaScript snippet to execute before returning content.
Return screenshot as base64-encoded PNG.
JSON scenario for custom headless browser actions.
Two-letter country code for geolocation (e.g., 'us').
Extraction rules (CSS selector or JSONPath).
Use premium proxy for scraping.
Use stealth (undetectable) proxy mode.
Block images and CSS resources on the page to speed up scraping.
CSS selector of element to screenshot.
Capture full-page screenshot instead of only viewport.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
SCRAPINGBEE_SCRAPING_BEE_PROXY_MODETool to fetch web content via ScrapingBee's Proxy Mode. Use when you need to route requests through ScrapingBee proxies with optional JS rendering and resource blocking.
Input parameters
The target URL to scrape through ScrapingBee Proxy Mode.
Cookies to send with the request as a key-value mapping.
Additional HTTP headers to forward to the target site. Each header will be prefixed with 'Spb-' and forwarded when forward_headers is enabled.
Request timeout in milliseconds.
Block ads and tracking scripts to speed up scraping.
Enable JavaScript rendering before returning content.
Session identifier (integer) to keep the same IP for multiple requests. Use the same number to maintain consistent IP across requests.
Custom JavaScript scenario name for advanced interactions.
Two-letter country code for geolocated proxy (e.g., 'us', 'fr').
Use premium proxies for higher reliability.
Use stealth proxy mode for extra undetectability.
Block images and CSS resources to speed up scraping. Only relevant when render_js is enabled.
Forward original request headers to the target site.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
SCRAPINGBEE_STEALTH_PROXYTool to perform stealth scraping via ScrapingBee's Stealth Proxy mode. Use when you encounter anti-bot measures requiring undetectable requests.
Input parameters
The URL of the webpage to retrieve using stealth proxy.
Wait time in milliseconds before returning the response.
Device type to emulate during rendering. Options: 'desktop' or 'mobile'.
Custom cookies in semicolon-separated format: 'name1=value1;name2=value2'.
Render JavaScript on the page before returning the response.
Two-letter country code for proxy geolocation (e.g., 'us', 'de').
Extraction rules in JSON string for structured data.
Use premium proxies for higher reliability.
Enable stealth proxy mode. Use when the target site blocks bots.
Block images, styles, and fonts for faster loads.
Forward original request headers from the browser.
Return the raw page source instead of text.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
SCRAPINGBEE_USAGE_STATSTool to retrieve usage statistics for your ScrapingBee account. Use when you need to monitor remaining credits and request count.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
Agents can call 5 Scrapingbee actions on Nagent, including "ScrapingBee Data Extraction", "ScrapingBee HTML Fetch" and "ScrapingBee Proxy Mode". ScrapingBee Data Extraction: Extract structured data from a webpage using CSS or XPath selectors. Each action is listed on this page with its input parameters and its output.
Scrapingbee connects with an API key, under your workspace's own connection. Nothing is enabled on connect: each action is allowed one at a time and can be scoped to the agents that need it.
ScrapingBee Data Extraction takes 3 required inputs: url, api_key and extractor. It also takes 7 optional inputs: wait, device, javascript, country_code, premium_proxy, block_resources and forward_headers. It returns data, error and successful.