Integration · Developer tools

Connect Claude & ChatGPT to Hyperbrowser

Ask your AI to read a web page, crawl a site or pull structured data out of a set of URLs, using your own Hyperbrowser account. It can search the web, scrape one page or many at once, and start an AI browser agent that clicks through a site to carry out a task. Longer jobs return a job ID, and the AI checks back for the result.

Connect Hyperbrowser free

Free plan, no credit card. Takes about a minute.

Try asking

“Get the content of this pricing page.”
The AI fetches the page directly and returns its content.
“Crawl our docs site, at most 50 pages.”
It starts a crawl job from the URL with a page limit and reads the collected pages as the job runs.
“From these five product pages, pull out name, price and stock status.”
It starts an extract job with your description, optionally with a JSON schema, and returns the structured data when it is done.

What your AI can do

Scrape and crawl

Fetch a single page, scrape one or many URLs, and crawl a site up to a page limit, with output options per job.

Extract and search

Pull structured data out of pages with a prompt and optional JSON schema, and search the web with filters.

Browser agents and sessions

Run a browser agent task in plain words, follow or stop it, stop cloud browser sessions and manage saved profiles.

“Search the web for MCP setup guides, only on dev.to.”
It runs a web search with a site filter and lists the results.
“Scrape these 20 URLs at once.”
It starts a batch scrape job and reads each page's content when the batch is finished.
“Open example.com, find the contact form and tell me which fields it asks for.”
It starts a browser agent task described in plain words, then reads the agent's steps and final result.

17 tools for Hyperbrowser

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

batch_scrape_pagesVisible to othersStart scraping a list of pages at once. Pass `urls` and optional `scrape_options`. Returns a batch job ID.
crawl_websiteChanges dataStart crawling a website from `url`, following links up to `max_pages`. Returns a crawl job ID.
create_profileChanges dataCreate a new saved browser profile, optionally with a `name`.
extract_dataChanges dataStart extracting structured data from one or more `urls`. Describe what to pull out in `prompt`, and optionally give a JSON `schema` for the result. Returns a job ID.
fetch_web_pageChanges dataFetch one web page's content directly and return it. Pass `url` and optionally `outputs` to choose formats and selectors.
get_batch_scrape_resultGet a batch scrape job's status and results. Pass `batch_id`.
get_browser_agent_taskGet a browser agent task's status, steps and final result. Pass `browser_use_id`.
get_crawl_resultGet a crawl job's status and the pages collected so far. Pass `crawl_id`, optionally `page` and `batch_size` to page through results.
get_extract_resultGet an extract job's status and the extracted data. Pass `extract_id`.
get_scrape_resultGet a scrape job's status and the page content once finished. Pass `scrape_id`.
get_sessionGet one cloud browser session's details and status by `session_id`.
list_profilesList saved browser profiles (kept logins and cookies), optionally filtered by `name`.
run_browser_agent_taskChanges dataStart an AI browser agent that carries out a task on the web, described in plain words in `task`. Optionally cap `max_steps`. Returns a task ID. The agent can click, type and submit forms on sites.
scrape_pageChanges dataStart scraping one web page. Pass `url` and optionally `scrape_options` (output formats, tags to include or exclude, wait time). Returns a job ID; fetch the content with get_scrape_result.
search_webChanges dataSearch the web for `query`. Optional `filters` (site, words in title or URL, excluded terms) and `location` narrow results.
stop_browser_agent_taskVisible to othersStop a running browser agent task. Pass `browser_use_id`.
stop_sessionVisible to othersStop a cloud browser session by `session_id`.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Hyperbrowser from the app list.
2
Paste your key
Paste your Hyperbrowser API key. Sign in to the Hyperbrowser dashboard at app.hyperbrowser.ai and copy your API key, or create a new one under API keys.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Hyperbrowser?

The AI can fetch and scrape web pages, run batch scrapes and crawls, extract structured data, search the web and start browser agent tasks. It can read the status and results of those jobs, stop an agent task or a cloud browser session, and list and create saved browser profiles.

Does the AI see my Hyperbrowser API key?

No. Your Hyperbrowser API key is encrypted at rest and never shown to the AI. After you save it, it is not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, for example scraping and search without browser agents. A browser agent can click, type and submit forms on real sites, so starting one can require your approval before it runs. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Hyperbrowser account.

Why do some jobs return an ID instead of a result?

Scrapes, batch scrapes, crawls, extracts and browser agent tasks run as jobs in Hyperbrowser. The AI starts the job, gets a job ID and then reads the status and results with the matching get tool. fetch_web_page is the exception and returns a page's content directly.