Skip to content

Connectors · Developer & infrastructure

Browserless MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Browserless and the rest of your systems. It gives your agent 64 Browserless actions (35 reads, 20 writes and 9 that wait for a confirmation) and holds the Browserless credential itself, so the agent never does. It paces Browserless's limits and records every call.

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

claude mcp add --transport http actionlake https://connect.actionlake.com/mcp
the first call
do {  action: "browserless.onepassword_integrations.list",  intent: "check which Browserless account is connected"}# List 1Password integrations · read · 1 upstream call
  • 64actionsin one signed file
  • 35readlook, never change
  • 20writecreate and update
  • 9wait for a yesdestructive, confirmed first

Set up

Connect Browserless in three steps.

  1. 01

    Add actionlake to Claude.

    Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.

    https://connect.actionlake.com/mcp
  2. 02

    Connect Browserless.

    Bearer token. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.

    • api_tokenBrowserless API token: browserless.io > Account dashboard > API token (self-hosted: the TOKEN environment variable)
    • hostWhere your Browserless lives: your host
    account { op: "connections.requirements", service: "browserless" }
    # free: exactly what to paste, and what is found for you
  3. 03

    Make the first call. Check before a write.

    Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

    do    { action: "browserless.onepassword_integrations.list" }# list 1password integrations · read · 1 upstream callcheck { action: "browserless.onepassword_integrations.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Browserless at a glance

Sign-in
Bearer token. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
You provide
api_token
Setup fields
host
Limits we pace to
60 calls a minute, per connection
API covered
Browserless Enterprise and Cloud REST API 2.57.3, at your Browserless instance's API base URL, which you enter when connecting
Built from
Browserless's documentation, read 23 October 2026docs.browserless.iodocs.browserless.io
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
29 of 35 reads, for up to an hour; a write clears what it affects
Shared entities
2, so ask can line Browserless up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Browserless.

64 actions in 24 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 35readlook, never change
  • 20writecreate and update
  • 9destructivewait for a confirmation
  1. Profiles

    9

    3 read · 4 write · 2 destructive

    List authentication profiles · Get an authentication profile · Start a profile-authoring session · Create a profile from captured state

  2. Chrome

    6

    3 read · 3 write · 0 destructive

    Run a BrowserQL query with Chrome · Export a page with Chrome · Run a function in Chrome · Run a Lighthouse audit with Chrome

  3. Chromium

    6

    4 read · 2 write · 0 destructive

    Run a BrowserQL query with Chromium · Export a page with Chromium · Run a function in Chromium · Run a Lighthouse audit with Chromium

  4. Agent runs

    4

    2 read · 1 write · 1 destructive

    Start an agent run · List agent runs · Get an agent run · Cancel or delete an agent run

  5. Crawls

    4

    2 read · 1 write · 1 destructive

    List crawl jobs · Start a crawl job · Get a crawl job and its pages · Cancel a crawl job

  6. Onepassword integrations

    4

    2 read · 1 write · 1 destructive

    List 1Password integrations · Register a 1Password integration · Get a 1Password integration · Delete a 1Password integration

  7. Devtools

    3

    2 read · 1 write · 0 destructive

    Open a new DevTools target · Get the DevTools protocol · Get DevTools version info

  8. Edge

    3

    2 read · 1 write · 0 destructive

    Run a function in Edge · Run a Lighthouse audit with Edge · Scrape elements with Edge

16 more areas · 25 actions
  • Live urls 3
  • Onepassword autologin 3
  • Sessions 3
  • Browsers 2
  • Scrapers 2
  • System 2
  • Browser pages 1
  • Onepassword audit 1
  • Proxy cities 1
  • Scrape 1
  • Site map 1
  • Skills 1
  • Smart scrape 1
  • Stealth 1
  • Unblock 1
  • Web search 1

Top actions

The Browserless API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 12 of Browserless's 12 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
browserless.onepassword_integrations.listList 1Password integrationsLists the 1Password integrations owned by the requesting token.readidentity.service_account.list35 connectors
browserless.profiles.listList authentication profilesLists authentication profiles owned by the requesting token.readidentity.user_session.list16 connectors
browserless.onepassword_integrations.createRegister a 1Password integrationRegisters a 1Password integration for the requesting token.writeidentity.service_account.create32 connectors
browserless.onepassword_integrations.deleteDelete a 1Password integrationDeletes a 1Password integration owned by the requesting token.destructiveidentity.service_account.delete25 connectors
browserless.profiles.getGet an authentication profileReturns metadata for a specific authentication profile.readidentity.user_session.get10 connectors
browserless.crawls.listList crawl jobsList all crawl jobs owned by the authenticated account (or, for keys not yet mapped to an account, the token).readautomation.background_job.list34 connectors
browserless.crawls.createStart a crawl jobStart an asynchronous crawl job that spiders a website and scrapes every discovered page.writeautomation.background_job.create29 connectors
browserless.web_search.searchSearch the web and scrape resultsSearch the web and optionally scrape result pages.readplatform.search_result.search18 connectors
browserless.system.activeCheck the service is upReturns a simple "204" HTTP code, with no response, indicating that the service itself is up and running.readplatform.health.get18 connectors
browserless.chrome.bqlRun a BrowserQL query with ChromeParses and executes BrowserQL requests, powered by the BrowserQL Editor or by other API integrations.writethis connector
browserless.scrape.runScrape elements from a pageA JSON-based API that returns text, html, and meta-data from a given list of selectors.readthis connector
browserless.edge.scrapeScrape elements with EdgeReturns text, HTML, and metadata from a given list of CSS selectors against the rendered DOM (Edge).readthis connector

Try it

Prompts to paste into Claude.

Each one names the Browserless actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    List 1Password integrations in Browserless and tell me what stands out.

    callsbrowserless.onepassword_integrations.list

  2. Writewrite

    Register a 1Password integration in Browserless. Show me exactly what you will send before you save it.

    callsbrowserless.onepassword_integrations.create

Governance for Browserless

Destructive actions wait for a yes.

9 Browserless actions are destructive. These are all of them:

  • Delete a 1Password integrationbrowserless.onepassword_integrations.delete
  • Cancel a crawl jobbrowserless.crawls.cancel
  • Kill browser sessionsbrowserless.browsers.kill
  • Terminate a browserbrowserless.browsers.delete
  • Delete an authentication profilebrowserless.profiles.delete
  • Delete a browser sessionbrowserless.sessions.delete
  • Close a live URLbrowserless.live_urls.delete
  • Cancel or delete an agent runbrowserless.agent_runs.delete
  • Remove sites from a profilebrowserless.profiles.remove_domains

A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

dobrowserless.onepassword_integrations.deleteconfirm_policy: human

Delete a 1Password integration?

intent: the user asked me to delete a 1Password integration

run suspended · waiting for a named person · key browserless-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "browserless-agent",  allowed_actions: ["browserless.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "browserless.onepassword_integrations.delete", approver: "[email protected]" },    { match: "browserless.crawls.cancel", approver: "[email protected]" },    { match: "browserless.browsers.kill", approver: "[email protected]" }  ]}
Illustrative. Scoped to Browserless only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Browserless costs on actionlake.

Every Browserless action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Browserless refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

  • 1upstream call per actionall 64 Browserless actions
  • 64actions pageeach page of a long list is one call
  • 29reads can be cacheda cache hit costs nothing
  • 100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Browserless action.

64 actions by area: the id your agent calls, what it does, and its effect.

Profiles9
  • browserless.profiles.createStart a profile-authoring sessionwrite
  • browserless.profiles.deleteDelete an authentication profiledestructive
  • browserless.profiles.download_stateDownload a profile's captured stateread
  • browserless.profiles.getGet an authentication profileread
  • browserless.profiles.listList authentication profilesread
  • browserless.profiles.refreshReplace a profile's captured statewrite
  • browserless.profiles.remove_domainsRemove sites from a profiledestructive
  • browserless.profiles.renameRename an authentication profilewrite
  • browserless.profiles.uploadCreate a profile from captured statewrite
Chrome6
  • browserless.chrome.bqlRun a BrowserQL query with Chromewrite
  • browserless.chrome.exportExport a page with Chromeread
  • browserless.chrome.functionRun a function in Chromewrite
  • browserless.chrome.performanceRun a Lighthouse audit with Chromeread
  • browserless.chrome.scrapeScrape elements with Chromeread
  • browserless.chrome.unblockUnblock a URL with Chromewrite
Chromium6
  • browserless.chromium.agent_capabilitiesGet Chromium agent route capabilitiesread
  • browserless.chromium.bqlRun a BrowserQL query with Chromiumwrite
  • browserless.chromium.exportExport a page with Chromiumread
  • browserless.chromium.functionRun a function in Chromiumwrite
  • browserless.chromium.performanceRun a Lighthouse audit with Chromiumread
  • browserless.chromium.scrapeScrape elements with Chromiumread
Agent runs4
  • browserless.agent_runs.createStart an agent runwrite
  • browserless.agent_runs.deleteCancel or delete an agent rundestructive
  • browserless.agent_runs.getGet an agent runread
  • browserless.agent_runs.listList agent runsread
Crawls4
  • browserless.crawls.cancelCancel a crawl jobdestructive
  • browserless.crawls.createStart a crawl jobwrite
  • browserless.crawls.getGet a crawl job and its pagesread
  • browserless.crawls.listList crawl jobsread
Onepassword integrations4
  • browserless.onepassword_integrations.createRegister a 1Password integrationwrite
  • browserless.onepassword_integrations.deleteDelete a 1Password integrationdestructive
  • browserless.onepassword_integrations.getGet a 1Password integrationread
  • browserless.onepassword_integrations.listList 1Password integrationsread
Devtools3
  • browserless.devtools.new_targetOpen a new DevTools targetwrite
  • browserless.devtools.protocolGet the DevTools protocolread
  • browserless.devtools.versionGet DevTools version inforead
Edge3
  • browserless.edge.functionRun a function in Edgewrite
  • browserless.edge.performanceRun a Lighthouse audit with Edgeread
  • browserless.edge.scrapeScrape elements with Edgeread
Live urls3
  • browserless.live_urls.createCreate a live URL for a browserwrite
  • browserless.live_urls.deleteClose a live URLdestructive
  • browserless.live_urls.getGet a live URL's statusread
Onepassword autologin3
  • browserless.onepassword_autologin.createStart a 1Password autologin taskwrite
  • browserless.onepassword_autologin.formGet autologin form choicesread
  • browserless.onepassword_autologin.getGet a 1Password autologin taskread
Sessions3
  • browserless.sessions.bqlRun a BrowserQL query in a sessionwrite
  • browserless.sessions.createCreate a browser sessionwrite
  • browserless.sessions.deleteDelete a browser sessiondestructive
Browsers2
  • browserless.browsers.deleteTerminate a browserdestructive
  • browserless.browsers.killKill browser sessionsdestructive
Scrapers2
  • browserless.scrapers.listList predefined scrapersread
  • browserless.scrapers.runRun a predefined scraperread
System2
  • browserless.system.activeCheck the service is upread
  • browserless.system.metaGet system versionsread
Browser pages1
  • browserless.browser_pages.listList a browser's open tabsread
Onepassword audit1
  • browserless.onepassword_audit.listList a 1Password integration's credential auditread
Proxy cities1
  • browserless.proxy_cities.getList proxy citiesread
Scrape1
  • browserless.scrape.runScrape elements from a pageread
Site map1
  • browserless.site_map.queryDiscover a site's URLsread
Skills1
  • browserless.skills.listList automation skills for a siteread
Smart scrape1
  • browserless.smart_scrape.runSmart-scrape a URLread
Stealth1
  • browserless.stealth.bqlRun a BrowserQL query with the stealth browserwrite
Unblock1
  • browserless.unblock.runUnblock a URL past bot detectionwrite
Web search1
  • browserless.web_search.searchSearch the web and scrape resultsread

Questions

The Browserless MCP server, answered.

Is there a Browserless MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 64 Browserless actions across profiles, chrome, chromium and agent runs, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Browserless.

What do I need to connect Browserless?

Browserless API token: browserless.io > Account dashboard > API token (self-hosted: the TOKEN environment variable). Plus your host. The free call account { op: "connections.requirements", service: "browserless" } lists exactly what to paste.

Can an AI agent delete or change things in Browserless?

Only if its key allows it, and only with an explicit confirmation. 9 Browserless actions are destructive, including delete a 1Password integration, cancel a crawl job and kill browser sessions. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my Browserless API limit?

actionlake paces every call to the limits in the Browserless service file: 60 calls a minute per connection. Repeated reads are answered from cache (29 Browserless actions can be), which never reaches Browserless.

What does a Browserless action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (64 Browserless actions page). Cache hits, searches, dry runs and calls Browserless refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

Which Browserless API does it cover?

The Browserless Enterprise and Cloud REST API 2.57.3 at your Browserless instance's API base URL, which you enter when connecting. It is built from Browserless's documentation as read on 23 October 2026. 64 actions in 24 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Browserless data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to an hour, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Browserless credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

Which AI clients can use it?

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

Give your agent Browserless today.

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.