MCP Server

oassis — web scraping, crawling and browser control for AI agents

dev.oassis/scraper-crawler
Developer Tools Search & Research Public & reachable MCP 2026-07-28

What this MCP does

Searches, scrapes, crawls, maps, and interacts with public websites using browser sessions, structured extraction, screenshots, and accessibility data.

web_act
Runs actions against an open session and returns the resulting state. Actions: {navigate}, {click:{ref}}, {type:{ref,text,clear}}, {select:{ref,value}}, {press}, {scroll}, {wait}, {back}. The `ref` is the one `controls` gave you. It stops at the first failure and tells you where. $0.0005 per action. Reference for the oassis API: https://oassis.dev/openapi.json
Open world
Input schema
{'type': 'object', 'required': ['sessionId', 'actions'], 'properties': {'pdf': {'type': 'object', 'properties': {'scale': {'type': 'number', 'maximum': 2, 'minimum': 0.1}, 'format': {'enum': ['letter', 'legal', 'tabloid', 'ledger', 'a0', 'a1', 'a2', 'a3', 'a4', 'a5', 'a6'], 'type': 'string'}, 'landscape': {'type': 'boolean'}, 'printBackground': {'type': 'boolean'}}, 'description': 'Options for the `pdf` format: `format` (a4, letter, legal…), `landscape`, `printBackground`, `scale` 0.1 to 2.', 'additionalProperties': False}, 'url': {'type': 'string', 'format': 'uri', 'description': 'Page to process.'}, 'json': {'type': 'object', 'properties': {'model': {'type': 'string', 'minLength': 1}, 'prompt': {'type': 'string', 'minLength': 1}, 'schema': {'type': 'object', 'additionalProperties': {}}}, 'description': 'For the `json` format: a `prompt` describing what to pull out, a JSON `schema` to fill, or both.', 'additionalProperties': False}, 'wait': {'type': 'object', 'properties': {'until': {'enum': ['load', 'domcontentloaded', 'networkidle0', 'networkidle2'], 'type': 'string'}, 'timeout': {'type': 'integer', 'maximum': 60000, 'minimum': 100}, 'selector': {'type': 'string', 'minLength': 1}}, 'description': 'When to consider the page loaded: `until` (load, domcontentloaded, networkidle0, networkidle2), `selector` to wait for, `timeout` in ms.', 'additionalProperties': False}, 'block': {'type': 'object', 'properties': {'urlPatterns': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}, 'resourceTypes': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}}, 'description': 'What not to load, which makes a render faster and cheaper. `resourceTypes` takes image, font, stylesheet, media… and `urlPatterns` are regular expressions, not globs: `\\.svg$`, not `*.svg`.', 'additionalProperties': False}, 'links': {'type': 'object', 'properties': {'visibleOnly': {'type': 'boolean'}, 'excludeExternal': {'type': 'boolean'}}, 'description': 'Options for the `links` format: `visibleOnly`, `excludeExternal` to keep only internal links.', 'additionalProperties': False}, 'maxAge': {'type': 'integer', 'maximum': 604800000, 'minimum': 0, 'description': 'Accept an answer up to this many milliseconds old. A cache hit costs $0.0002 instead of the format price. Leave it out to force a fresh render.'}, 'actions': {'type': 'array', 'items': {'anyOf': [{'type': 'object', 'required': ['navigate'], 'properties': {'navigate': {'type': 'string', 'format': 'uri'}}, 'additionalProperties': False}, {'type': 'object', 'required': ['click'], 'properties': {'click': {'type': 'object', 'properties': {'ref': {'type': 'string', 'minLength': 1}, 'selector': {'type': 'string', 'minLength': 1}}, 'additionalProperties': False}}, 'additionalProperties': False}, {'type': 'object', 'required': ['type'], 'properties': {'type': {'type': 'object', 'required': ['text'], 'properties': {'ref': {'type': 'string', 'minLength': 1}, 'text': {'type': 'string'}, 'clear': {'type': 'boolean'}, 'selector': {'type': 'string', 'minLength': 1}}, 'additionalProperties': False}}, 'additionalProperties': False}, {'type': 'object', 'required': ['select'], 'properties': {'select': {'type': 'object', 'required': ['value'], 'properties': {'ref': {'type': 'string', 'minLength': 1}, 'value': {'type': 'string'}, 'selector': {'type': 'string', 'minLength': 1}}, 'additionalProperties': False}}, 'additionalProperties': False}, {'type': 'object', 'required': ['press'], 'properties': {'press': {'type': 'string', 'minLength': 1}}, 'additionalProperties': False}, {'type': 'object', 'required': ['scroll'], 'properties': {'scroll': {'type': 'object', 'properties': {'by': {'type': 'number'}, 'to': {'enum': ['top', 'bottom'], 'type': 'string'}}, 'additionalProperties': False}}, 'additionalProperties': False}, {'type': 'object', 'required': ['wait'], 'properties': {'wait': {'type': 'object', 'properties': {'ms': {'type': 'integer', 'maximum': 30000, 'minimum': 0}, 'selector': {'type': 'string', 'minLength': 1}}, 'additionalProperties': False}}, 'additionalProperties': False}, {'type': 'object', 'required': ['back'], 'properties': {'back': {'type': 'boolean', 'const': True}}, 'additionalProperties': False}]}, 'maxItems': 25, 'minItems': 1, 'description': 'What to do, in order: click, type, select, scroll, hover, press, navigate, wait. One to twenty-five per call.'}, 'formats': {'type': 'array', 'items': {'enum': ['html', 'markdown', 'links', 'screenshot', 'pdf', 'elements', 'json', 'accessibility', 'controls'], 'type': 'string'}, 'default': ['controls'], 'maxItems': 9, 'minItems': 1, 'description': 'Outputs you want in the same response. `controls` is the map of what can be clicked; `elements` needs `selectors`; `json` needs `json.prompt` or `json.schema`.'}, 'request': {'type': 'object', 'properties': {'auth': {'type': 'object', 'required': ['username', 'password'], 'properties': {'password': {'type': 'string'}, 'username': {'type': 'string'}}, 'additionalProperties': False}, 'cookies': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'value'], 'properties': {'name': {'type': 'string', 'minLength': 1}, 'path': {'type': 'string'}, 'value': {'type': 'string'}, 'domain': {'type': 'string'}, 'secure': {'type': 'boolean'}, 'httpOnly': {'type': 'boolean'}}, 'additionalProperties': False}}, 'headers': {'type': 'object', 'additionalProperties': {'type': 'string'}}, 'userAgent': {'type': 'string', 'minLength': 1}}, 'description': 'How to make the request: `headers`, `cookies`, `userAgent`, and `auth` for basic authentication.', 'additionalProperties': False}, 'binaryAs': {'enum': ['base64', 'url'], 'type': 'string', 'default': 'base64', 'description': 'How a screenshot or a PDF comes back. `base64` is the only value today.'}, 'controls': {'type': 'object', 'properties': {'limit': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'visibleOnly': {'type': 'boolean'}}, 'description': 'Options for the `controls` format: `visibleOnly`, and `limit` (1 to 1000) to cut the map.', 'additionalProperties': False}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'description': 'The window to render in: `width`, `height`, `deviceScaleFactor`.', 'additionalProperties': False}, 'selectors': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'minItems': 1, 'description': 'CSS selectors for the `elements` format.'}, 'sessionId': {'type': 'string', 'minLength': 1, 'description': 'The session to act on, from `web_session_open`.'}, 'screenshot': {'type': 'object', 'properties': {'type': {'enum': ['png', 'jpeg', 'webp'], 'type': 'string'}, 'quality': {'type': 'integer', 'maximum': 100, 'minimum': 1}, 'fullPage': {'type': 'boolean'}, 'selector': {'type': 'string', 'minLength': 1}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'additionalProperties': False}, 'omitBackground': {'type': 'boolean'}}, 'description': 'Options for the `screenshot` format: `fullPage` for the whole page, `type` (png, jpeg, webp), `quality` 1-100 (not with png), `omitBackground`, `selector` to capture one element, and its own `viewport`.', 'additionalProperties': False}}}
web_batch_status
Checks a batch of scrapes: status, how many are done, and the results. Free. Pass `cancel: true` to stop it and get the urls it never read refunded.
Idempotent
Input schema
{'type': 'object', 'required': ['jobId'], 'properties': {'jobId': {'type': 'string', 'description': 'The batch, from web_scrape_batch.'}, 'limit': {'type': 'number', 'description': 'Results to return (default 50).'}, 'cancel': {'type': 'boolean', 'description': 'Stop the batch and refund what it did not read.'}}}
web_crawl
Follows a site's links and reads every page. Returns a jobId; poll it with web_crawl_status. Charged up front for the pages it is allowed to read (`limit`), and the pages it never reads are refunded. Use web_map first if you only need the urls. Reference for the oassis API: https://oassis.dev/openapi.json
Read only Open world
Input schema
{'type': 'object', 'required': ['url'], 'properties': {'url': {'type': 'string', 'format': 'uri', 'description': 'Page to process.'}, 'json': {'type': 'object', 'properties': {'model': {'type': 'string', 'minLength': 1}, 'prompt': {'type': 'string', 'minLength': 1}, 'schema': {'type': 'object', 'additionalProperties': {}}}, 'description': 'For the `json` format: a `prompt` describing what to pull out, a JSON `schema` to fill, or both.', 'additionalProperties': False}, 'wait': {'type': 'object', 'properties': {'until': {'enum': ['load', 'domcontentloaded', 'networkidle0', 'networkidle2'], 'type': 'string'}, 'timeout': {'type': 'integer', 'maximum': 60000, 'minimum': 100}, 'selector': {'type': 'string', 'minLength': 1}}, 'description': 'When to consider the page loaded: `until` (load, domcontentloaded, networkidle0, networkidle2), `selector` to wait for, `timeout` in ms.', 'additionalProperties': False}, 'block': {'type': 'object', 'properties': {'urlPatterns': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}, 'resourceTypes': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}}, 'description': 'What not to load, which makes a render faster and cheaper. `resourceTypes` takes image, font, stylesheet, media… and `urlPatterns` are regular expressions, not globs: `\\.svg$`, not `*.svg`.', 'additionalProperties': False}, 'limit': {'type': 'integer', 'default': 25, 'maximum': 200, 'minimum': 1, 'description': 'Pages this crawl may read, up to 200. Charged up front; what it does not read comes back.'}, 'links': {'type': 'object', 'properties': {'visibleOnly': {'type': 'boolean'}, 'excludeExternal': {'type': 'boolean'}}, 'description': 'Options for the `links` format: `visibleOnly`, `excludeExternal` to keep only internal links.', 'additionalProperties': False}, 'maxAge': {'type': 'integer', 'maximum': 604800000, 'minimum': 0, 'description': 'Accept an answer up to this many milliseconds old. A cache hit costs $0.0002 instead of the format price. Leave it out to force a fresh render.'}, 'formats': {'type': 'array', 'items': {'enum': ['html', 'markdown', 'links', 'elements', 'json', 'accessibility', 'controls'], 'type': 'string'}, 'description': 'Outputs you want in the same response. `controls` is the map of what can be clicked; `elements` needs `selectors`; `json` needs `json.prompt` or `json.schema`. A screenshot or a PDF is not available here: every result is kept until you collect it. Ask for those one url at a time on web_scrape.'}, 'request': {'type': 'object', 'properties': {'auth': {'type': 'object', 'required': ['username', 'password'], 'properties': {'password': {'type': 'string'}, 'username': {'type': 'string'}}, 'additionalProperties': False}, 'cookies': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'value'], 'properties': {'name': {'type': 'string', 'minLength': 1}, 'path': {'type': 'string'}, 'value': {'type': 'string'}, 'domain': {'type': 'string'}, 'secure': {'type': 'boolean'}, 'httpOnly': {'type': 'boolean'}}, 'additionalProperties': False}}, 'headers': {'type': 'object', 'additionalProperties': {'type': 'string'}}, 'userAgent': {'type': 'string', 'minLength': 1}}, 'description': 'How to make the request: `headers`, `cookies`, `userAgent`, and `auth` for basic authentication.', 'additionalProperties': False}, 'binaryAs': {'enum': ['base64', 'url'], 'type': 'string', 'default': 'base64', 'description': 'How a screenshot or a PDF comes back. `base64` is the only value today.'}, 'controls': {'type': 'object', 'properties': {'limit': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'visibleOnly': {'type': 'boolean'}}, 'description': 'Options for the `controls` format: `visibleOnly`, and `limit` (1 to 1000) to cut the map.', 'additionalProperties': False}, 'maxDepth': {'type': 'integer', 'default': 2, 'maximum': 5, 'minimum': 0, 'description': 'How far from the starting page to follow links. 0 reads only the starting page.'}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'description': 'The window to render in: `width`, `height`, `deviceScaleFactor`.', 'additionalProperties': False}, 'selectors': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'minItems': 1, 'description': 'CSS selectors for the `elements` format.'}, 'excludePaths': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Discard urls whose path contains one of these. Up to twenty. It wins over `includePaths`.'}, 'includePaths': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Keep only urls whose path contains one of these. Up to twenty.'}, 'includeSubdomains': {'type': 'boolean', 'description': 'Treat subdomains of the same site as part of it.'}}}
web_crawl_status
Checks a crawl: status, pages read, discovered and still queued, and the pages themselves. Free. Pass `cancel: true` to stop it and get the unread pages refunded.
Idempotent
Input schema
{'type': 'object', 'required': ['jobId'], 'properties': {'jobId': {'type': 'string', 'description': 'The crawl, from web_crawl.'}, 'limit': {'type': 'number', 'description': 'Pages to return (default 50).'}, 'cancel': {'type': 'boolean', 'description': 'Stop the crawl and refund what it did not read.'}}}
web_feedback
Tell us an answer was good or bad. FREE. Use it when a result is wrong — empty markdown, a control map missing a button, data that does not match the page — with the url or the jobId so it can be reproduced. It is the only way we learn that we read a page badly: our logs cannot tell that apart from a page that is simply like that. Reference for the oassis API: https://oassis.dev/openapi.json
Input schema
{'type': 'object', 'required': ['verdict'], 'properties': {'url': {'type': 'string', 'format': 'uri', 'description': 'The page that came out wrong: usually the fastest way to reproduce it.'}, 'route': {'type': 'string', 'maxLength': 120, 'minLength': 1, 'description': 'Which tool or endpoint it is about, as you called it.'}, 'comment': {'type': 'string', 'maxLength': 2000, 'minLength': 1, 'description': 'What you expected and what you got.'}, 'verdict': {'enum': ['good', 'bad'], 'type': 'string', 'description': 'Did the answer do the job or not.'}, 'reference': {'type': 'string', 'maxLength': 200, 'minLength': 1, 'description': 'The jobId or sessionId it happened on, so it can be looked up.'}}}
web_map
Every url of a site, fast and cheap: its sitemap plus, optionally, the links on the page. Use it BEFORE crawling, to see what is there and decide what is worth reading. $0.001 with `includePage: false` (no browser at all), $0.0015 with the page. Reference for the oassis API: https://oassis.dev/openapi.json
Read only Open world Idempotent
Input schema
{'type': 'object', 'required': ['url'], 'properties': {'url': {'type': 'string', 'format': 'uri', 'description': 'Page to process.'}, 'limit': {'type': 'integer', 'default': 1000, 'maximum': 5000, 'minimum': 1, 'description': 'Urls to return. Up to 5000.'}, 'search': {'type': 'string', 'maxLength': 200, 'minLength': 1, 'description': 'Keep only urls containing this text.'}, 'includePage': {'type': 'boolean', 'description': 'Render the page too, to catch urls the sitemap does not list. `false` uses no browser at all and costs a third.'}, 'excludePaths': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Discard urls whose path contains one of these. Up to twenty. It wins over `includePaths`.'}, 'includePaths': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Keep only urls whose path contains one of these. Up to twenty.'}, 'includeSubdomains': {'type': 'boolean', 'description': 'Treat subdomains of the same site as part of it.'}}}
web_scrape
Processes a page and returns every output you ask for at once: markdown, html, links, screenshot, PDF, accessibility tree, elements by selector, AI-structured data, and `controls` (what can be clicked). One call, and a partial failure does not void the rest. From $0.001 per output. A url pointing at a PDF, Word, Excel or CSV file is converted to markdown instead, with no browser, for $0.002. Reference for the oassis API: https://oassis.dev/openapi.json
Read only Open world
Input schema
{'type': 'object', 'properties': {'pdf': {'type': 'object', 'properties': {'scale': {'type': 'number', 'maximum': 2, 'minimum': 0.1}, 'format': {'enum': ['letter', 'legal', 'tabloid', 'ledger', 'a0', 'a1', 'a2', 'a3', 'a4', 'a5', 'a6'], 'type': 'string'}, 'landscape': {'type': 'boolean'}, 'printBackground': {'type': 'boolean'}}, 'description': 'Options for the `pdf` format: `format` (a4, letter, legal…), `landscape`, `printBackground`, `scale` 0.1 to 2.', 'additionalProperties': False}, 'url': {'type': 'string', 'format': 'uri', 'description': 'Page to process.'}, 'html': {'type': 'string', 'minLength': 1, 'description': 'Raw HTML instead of `url`, for a page you already have. Nothing leaves for the network.'}, 'json': {'type': 'object', 'properties': {'model': {'type': 'string', 'minLength': 1}, 'prompt': {'type': 'string', 'minLength': 1}, 'schema': {'type': 'object', 'additionalProperties': {}}}, 'description': 'For the `json` format: a `prompt` describing what to pull out, a JSON `schema` to fill, or both.', 'additionalProperties': False}, 'wait': {'type': 'object', 'properties': {'until': {'enum': ['load', 'domcontentloaded', 'networkidle0', 'networkidle2'], 'type': 'string'}, 'timeout': {'type': 'integer', 'maximum': 60000, 'minimum': 100}, 'selector': {'type': 'string', 'minLength': 1}}, 'description': 'When to consider the page loaded: `until` (load, domcontentloaded, networkidle0, networkidle2), `selector` to wait for, `timeout` in ms.', 'additionalProperties': False}, 'block': {'type': 'object', 'properties': {'urlPatterns': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}, 'resourceTypes': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}}, 'description': 'What not to load, which makes a render faster and cheaper. `resourceTypes` takes image, font, stylesheet, media… and `urlPatterns` are regular expressions, not globs: `\\.svg$`, not `*.svg`.', 'additionalProperties': False}, 'links': {'type': 'object', 'properties': {'visibleOnly': {'type': 'boolean'}, 'excludeExternal': {'type': 'boolean'}}, 'description': 'Options for the `links` format: `visibleOnly`, `excludeExternal` to keep only internal links.', 'additionalProperties': False}, 'maxAge': {'type': 'integer', 'maximum': 604800000, 'minimum': 0, 'description': 'Accept an answer up to this many milliseconds old. A cache hit costs $0.0002 instead of the format price. Leave it out to force a fresh render.'}, 'formats': {'type': 'array', 'items': {'enum': ['html', 'markdown', 'links', 'screenshot', 'pdf', 'elements', 'json', 'accessibility', 'controls'], 'type': 'string'}, 'default': ['markdown'], 'maxItems': 9, 'minItems': 1, 'description': 'Outputs you want in the same response. `controls` is the map of what can be clicked; `elements` needs `selectors`; `json` needs `json.prompt` or `json.schema`.'}, 'request': {'type': 'object', 'properties': {'auth': {'type': 'object', 'required': ['username', 'password'], 'properties': {'password': {'type': 'string'}, 'username': {'type': 'string'}}, 'additionalProperties': False}, 'cookies': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'value'], 'properties': {'name': {'type': 'string', 'minLength': 1}, 'path': {'type': 'string'}, 'value': {'type': 'string'}, 'domain': {'type': 'string'}, 'secure': {'type': 'boolean'}, 'httpOnly': {'type': 'boolean'}}, 'additionalProperties': False}}, 'headers': {'type': 'object', 'additionalProperties': {'type': 'string'}}, 'userAgent': {'type': 'string', 'minLength': 1}}, 'description': 'How to make the request: `headers`, `cookies`, `userAgent`, and `auth` for basic authentication.', 'additionalProperties': False}, 'binaryAs': {'enum': ['base64', 'url'], 'type': 'string', 'default': 'base64', 'description': 'How a screenshot or a PDF comes back. `base64` is the only value today.'}, 'controls': {'type': 'object', 'properties': {'limit': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'visibleOnly': {'type': 'boolean'}}, 'description': 'Options for the `controls` format: `visibleOnly`, and `limit` (1 to 1000) to cut the map.', 'additionalProperties': False}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'description': 'The window to render in: `width`, `height`, `deviceScaleFactor`.', 'additionalProperties': False}, 'selectors': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'minItems': 1, 'description': 'CSS selectors for the `elements` format.'}, 'screenshot': {'type': 'object', 'properties': {'type': {'enum': ['png', 'jpeg', 'webp'], 'type': 'string'}, 'quality': {'type': 'integer', 'maximum': 100, 'minimum': 1}, 'fullPage': {'type': 'boolean'}, 'selector': {'type': 'string', 'minLength': 1}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'additionalProperties': False}, 'omitBackground': {'type': 'boolean'}}, 'description': 'Options for the `screenshot` format: `fullPage` for the whole page, `type` (png, jpeg, webp), `quality` 1-100 (not with png), `omitBackground`, `selector` to capture one element, and its own `viewport`.', 'additionalProperties': False}}}
web_scrape_batch
A batch OF SCRAPES: reads a list of urls YOU give it (2 to 50, from any sites) and returns a jobId. It discovers nothing on its own — for that use web_crawl. Charged up front per url; urls that fail and urls served from the cache are refunded. Poll it with web_batch_status.
Read only Open world
Input schema
{'type': 'object', 'required': ['urls'], 'properties': {'json': {'type': 'object', 'properties': {'model': {'type': 'string', 'minLength': 1}, 'prompt': {'type': 'string', 'minLength': 1}, 'schema': {'type': 'object', 'additionalProperties': {}}}, 'description': 'For the `json` format: a `prompt` describing what to pull out, a JSON `schema` to fill, or both.', 'additionalProperties': False}, 'urls': {'type': 'array', 'items': {'type': 'string', 'format': 'uri'}, 'maxItems': 50, 'minItems': 2, 'description': 'The pages of this batch. Two to fifty, no duplicates: a batch charges per url.'}, 'wait': {'type': 'object', 'properties': {'until': {'enum': ['load', 'domcontentloaded', 'networkidle0', 'networkidle2'], 'type': 'string'}, 'timeout': {'type': 'integer', 'maximum': 60000, 'minimum': 100}, 'selector': {'type': 'string', 'minLength': 1}}, 'description': 'When to consider the page loaded: `until` (load, domcontentloaded, networkidle0, networkidle2), `selector` to wait for, `timeout` in ms.', 'additionalProperties': False}, 'block': {'type': 'object', 'properties': {'urlPatterns': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}, 'resourceTypes': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}}, 'description': 'What not to load, which makes a render faster and cheaper. `resourceTypes` takes image, font, stylesheet, media… and `urlPatterns` are regular expressions, not globs: `\\.svg$`, not `*.svg`.', 'additionalProperties': False}, 'links': {'type': 'object', 'properties': {'visibleOnly': {'type': 'boolean'}, 'excludeExternal': {'type': 'boolean'}}, 'description': 'Options for the `links` format: `visibleOnly`, `excludeExternal` to keep only internal links.', 'additionalProperties': False}, 'maxAge': {'type': 'integer', 'maximum': 604800000, 'minimum': 0, 'description': 'Accept an answer up to this many milliseconds old. A cache hit costs $0.0002 instead of the format price. Leave it out to force a fresh render.'}, 'formats': {'type': 'array', 'items': {'enum': ['html', 'markdown', 'links', 'elements', 'json', 'accessibility', 'controls'], 'type': 'string'}, 'description': 'Outputs you want in the same response. `controls` is the map of what can be clicked; `elements` needs `selectors`; `json` needs `json.prompt` or `json.schema`. A screenshot or a PDF is not available here: every result is kept until you collect it. Ask for those one url at a time on web_scrape.'}, 'request': {'type': 'object', 'properties': {'auth': {'type': 'object', 'required': ['username', 'password'], 'properties': {'password': {'type': 'string'}, 'username': {'type': 'string'}}, 'additionalProperties': False}, 'cookies': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'value'], 'properties': {'name': {'type': 'string', 'minLength': 1}, 'path': {'type': 'string'}, 'value': {'type': 'string'}, 'domain': {'type': 'string'}, 'secure': {'type': 'boolean'}, 'httpOnly': {'type': 'boolean'}}, 'additionalProperties': False}}, 'headers': {'type': 'object', 'additionalProperties': {'type': 'string'}}, 'userAgent': {'type': 'string', 'minLength': 1}}, 'description': 'How to make the request: `headers`, `cookies`, `userAgent`, and `auth` for basic authentication.', 'additionalProperties': False}, 'binaryAs': {'enum': ['base64', 'url'], 'type': 'string', 'default': 'base64', 'description': 'How a screenshot or a PDF comes back. `base64` is the only value today.'}, 'controls': {'type': 'object', 'properties': {'limit': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'visibleOnly': {'type': 'boolean'}}, 'description': 'Options for the `controls` format: `visibleOnly`, and `limit` (1 to 1000) to cut the map.', 'additionalProperties': False}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'description': 'The window to render in: `width`, `height`, `deviceScaleFactor`.', 'additionalProperties': False}, 'selectors': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'minItems': 1, 'description': 'CSS selectors for the `elements` format.'}}}
web_search_exa
Search the web with Exa's index: a query instead of a url, for when you do not know where to look. Returns title, url and a snippet per result. To read the pages, pass the urls to web_scrape_batch. The engine is named because the price is Exa's, passed through with no markup and read from its own payment challenge on every call — today $0.007 per search.
Read only Open world
Input schema
{'type': 'object', 'required': ['query'], 'properties': {'limit': {'type': 'integer', 'default': 10, 'maximum': 50, 'minimum': 1, 'description': 'Results to return. Up to 50.'}, 'query': {'type': 'string', 'maxLength': 500, 'minLength': 1, 'description': 'What to search for.'}, 'since': {'type': 'string', 'maxLength': 30, 'minLength': 4, 'description': 'Only results published after this ISO date.'}, 'domains': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Only these domains, subdomains included: `example.com` also matches `docs.example.com`.'}, 'snippets': {'type': 'boolean', 'description': 'Text alongside each result. On by default: a bare url is rarely enough.'}, 'excludeDomains': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'maxItems': 20, 'description': 'Never these domains, subdomains included.'}}}
web_session_close
Closes a session and stops billing browser time. Free. If you do not close it, it closes itself after a minute without use.
Idempotent
Input schema
{'type': 'object', 'required': ['sessionId'], 'properties': {'sessionId': {'type': 'string', 'description': 'The session to close, from web_session_open.'}}}
web_session_open
Opens a browser on a page and leaves it open, returning the map of controls. Use it when something has to be FILLED IN or CLICKED, not just read: inside the session the `controls` references keep working and you can act on the same state. $0.005 plus the outputs. Close it with web_session_close when you are done. Reference for the oassis API: https://oassis.dev/openapi.json
Open world
Input schema
{'type': 'object', 'required': ['url'], 'properties': {'pdf': {'type': 'object', 'properties': {'scale': {'type': 'number', 'maximum': 2, 'minimum': 0.1}, 'format': {'enum': ['letter', 'legal', 'tabloid', 'ledger', 'a0', 'a1', 'a2', 'a3', 'a4', 'a5', 'a6'], 'type': 'string'}, 'landscape': {'type': 'boolean'}, 'printBackground': {'type': 'boolean'}}, 'description': 'Options for the `pdf` format: `format` (a4, letter, legal…), `landscape`, `printBackground`, `scale` 0.1 to 2.', 'additionalProperties': False}, 'url': {'type': 'string', 'format': 'uri', 'description': 'Page to process.'}, 'json': {'type': 'object', 'properties': {'model': {'type': 'string', 'minLength': 1}, 'prompt': {'type': 'string', 'minLength': 1}, 'schema': {'type': 'object', 'additionalProperties': {}}}, 'description': 'For the `json` format: a `prompt` describing what to pull out, a JSON `schema` to fill, or both.', 'additionalProperties': False}, 'wait': {'type': 'object', 'properties': {'until': {'enum': ['load', 'domcontentloaded', 'networkidle0', 'networkidle2'], 'type': 'string'}, 'timeout': {'type': 'integer', 'maximum': 60000, 'minimum': 100}, 'selector': {'type': 'string', 'minLength': 1}}, 'description': 'When to consider the page loaded: `until` (load, domcontentloaded, networkidle0, networkidle2), `selector` to wait for, `timeout` in ms.', 'additionalProperties': False}, 'block': {'type': 'object', 'properties': {'urlPatterns': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}, 'resourceTypes': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}}}, 'description': 'What not to load, which makes a render faster and cheaper. `resourceTypes` takes image, font, stylesheet, media… and `urlPatterns` are regular expressions, not globs: `\\.svg$`, not `*.svg`.', 'additionalProperties': False}, 'links': {'type': 'object', 'properties': {'visibleOnly': {'type': 'boolean'}, 'excludeExternal': {'type': 'boolean'}}, 'description': 'Options for the `links` format: `visibleOnly`, `excludeExternal` to keep only internal links.', 'additionalProperties': False}, 'maxAge': {'type': 'integer', 'maximum': 604800000, 'minimum': 0, 'description': 'Accept an answer up to this many milliseconds old. A cache hit costs $0.0002 instead of the format price. Leave it out to force a fresh render.'}, 'formats': {'type': 'array', 'items': {'enum': ['html', 'markdown', 'links', 'screenshot', 'pdf', 'elements', 'json', 'accessibility', 'controls'], 'type': 'string'}, 'default': ['controls'], 'maxItems': 9, 'minItems': 1, 'description': 'Outputs you want in the same response. `controls` is the map of what can be clicked; `elements` needs `selectors`; `json` needs `json.prompt` or `json.schema`.'}, 'request': {'type': 'object', 'properties': {'auth': {'type': 'object', 'required': ['username', 'password'], 'properties': {'password': {'type': 'string'}, 'username': {'type': 'string'}}, 'additionalProperties': False}, 'cookies': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'value'], 'properties': {'name': {'type': 'string', 'minLength': 1}, 'path': {'type': 'string'}, 'value': {'type': 'string'}, 'domain': {'type': 'string'}, 'secure': {'type': 'boolean'}, 'httpOnly': {'type': 'boolean'}}, 'additionalProperties': False}}, 'headers': {'type': 'object', 'additionalProperties': {'type': 'string'}}, 'userAgent': {'type': 'string', 'minLength': 1}}, 'description': 'How to make the request: `headers`, `cookies`, `userAgent`, and `auth` for basic authentication.', 'additionalProperties': False}, 'binaryAs': {'enum': ['base64', 'url'], 'type': 'string', 'default': 'base64', 'description': 'How a screenshot or a PDF comes back. `base64` is the only value today.'}, 'controls': {'type': 'object', 'properties': {'limit': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'visibleOnly': {'type': 'boolean'}}, 'description': 'Options for the `controls` format: `visibleOnly`, and `limit` (1 to 1000) to cut the map.', 'additionalProperties': False}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'description': 'The window to render in: `width`, `height`, `deviceScaleFactor`.', 'additionalProperties': False}, 'selectors': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'minItems': 1, 'description': 'CSS selectors for the `elements` format.'}, 'screenshot': {'type': 'object', 'properties': {'type': {'enum': ['png', 'jpeg', 'webp'], 'type': 'string'}, 'quality': {'type': 'integer', 'maximum': 100, 'minimum': 1}, 'fullPage': {'type': 'boolean'}, 'selector': {'type': 'string', 'minLength': 1}, 'viewport': {'type': 'object', 'required': ['width', 'height'], 'properties': {'width': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'height': {'type': 'integer', 'maximum': 4096, 'minimum': 1}, 'deviceScaleFactor': {'type': 'number', 'maximum': 3, 'minimum': 0.1}}, 'additionalProperties': False}, 'omitBackground': {'type': 'boolean'}}, 'description': 'Options for the `screenshot` format: `fullPage` for the whole page, `type` (png, jpeg, webp), `quality` 1-100 (not with png), `omitBackground`, `selector` to capture one element, and its own `viewport`.', 'additionalProperties': False}}}
Changed
web_session_close
Oct. 5, 2026, 2:42 a.m.
Changed
web_feedback
Oct. 5, 2026, 2:42 a.m.
Changed
web_search_exa
Oct. 5, 2026, 2:42 a.m.
Changed
web_crawl_status
Oct. 5, 2026, 2:42 a.m.
Changed
web_crawl
Oct. 5, 2026, 2:42 a.m.
Changed
web_map
Oct. 5, 2026, 2:42 a.m.
Changed
web_batch_status
Oct. 5, 2026, 2:42 a.m.
Changed
web_scrape_batch
Oct. 5, 2026, 2:42 a.m.
Changed
web_act
Oct. 5, 2026, 2:42 a.m.
Changed
web_session_open
Oct. 5, 2026, 2:42 a.m.
Changed
web_scrape
Oct. 5, 2026, 2:42 a.m.
Changed
web_session_close
Oct. 3, 2026, 2:48 a.m.
Changed
web_feedback
Oct. 3, 2026, 2:48 a.m.
Changed
web_search_exa
Oct. 3, 2026, 2:48 a.m.
Changed
web_crawl_status
Oct. 3, 2026, 2:48 a.m.
Changed
web_crawl
Oct. 3, 2026, 2:48 a.m.
Changed
web_map
Oct. 3, 2026, 2:48 a.m.
Changed
web_batch_status
Oct. 3, 2026, 2:48 a.m.
Changed
web_scrape_batch
Oct. 3, 2026, 2:48 a.m.
Changed
web_act
Oct. 3, 2026, 2:48 a.m.
Changed
web_session_open
Oct. 3, 2026, 2:48 a.m.
Changed
web_scrape
Oct. 3, 2026, 2:48 a.m.
Added
web_session_close
Oct. 1, 2026, 2:40 a.m.
Added
web_feedback
Oct. 1, 2026, 2:40 a.m.
Added
web_search_exa
Oct. 1, 2026, 2:40 a.m.
Added
web_crawl_status
Oct. 1, 2026, 2:40 a.m.
Added
web_crawl
Oct. 1, 2026, 2:40 a.m.
Added
web_map
Oct. 1, 2026, 2:40 a.m.
Added
web_batch_status
Oct. 1, 2026, 2:40 a.m.
Added
web_scrape_batch
Oct. 1, 2026, 2:40 a.m.