Get started

BrowserAct for AI agents

BrowserAct runs hosted browser automation workflows that teams use to execute tasks against websites and track their status through to completion. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.

Connect BrowserActSee the 11 actions ↓
Web data and scrapingAPI keyMCP + RESTbrowseract.com
AUDIT LOG · BROWSERACTPOLICY: acme-support
09:41:07 · claude · u_8f2read
browseract.list_tasks
List Tasks✓ allowed · 212ms
09:41:08 · claude · u_8f2read
browseract.get_task
Get Task✓ allowed · 164ms
09:41:09 · claude · u_8f2write
browseract.resume_task
Resume Task✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
browseract.cancel_task
Cancel Task✕ blocked · policy: deny
EVERY BROWSERACT CALL, ON THE RECORD
01 · USE CASES

What agents do in BrowserAct.

01

Check a task's status

Read-only lookup of a task's status and results before deciding whether to resume or cancel it.

02

Run an automation workflow

Run a workflow or workflow template against a target after the user confirms the input parameters.

03

Require approval before cancelling a task

Treat cancelling a running task as needing confirmation since it stops in-progress automation.

02 · ACTIONS

11 BrowserAct actions, graded by risk.

Every BrowserAct action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

7

Look things up. Allowed by default.

  • browseract.get_task
    Get Task
  • browseract.list_tasks
    List Tasks
  • browseract.get_workflow
    Get Workflow
  • browseract.list_workflows
    List Workflows
  • browseract.list_proxy_regions
    List Proxy Regions
  • browseract.get_workflow_template
    Get Workflow Template
  • browseract.list_workflow_templates
    List Workflow Templates

write

3

Create and change things. Allow, or ask the user first.

  • browseract.resume_task
    Resume Task
  • browseract.run_workflow
    Run Workflow
  • browseract.run_workflow_template
    Run Workflow Template

destructive

1

Delete, cancel or archive. Ask first, or deny outright.

  • browseract.cancel_task
    Cancel Task
03 · HOW IT WORKS

BrowserAct in three steps.

  1. 01Your users connect BrowserActThey add their BrowserAct api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, writes like “resume Task” can wait for the user, and “cancel Task” can be denied outright.
  3. 03Any agent can actYour agent calls BrowserAct through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('browseract', {
  read: 'allow',
  write: 'ask',        // resume_task
  destructive: 'deny',  // cancel_task
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How BrowserAct connects.

Users add their BrowserAct api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

The same BrowserAct connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →

AUTH
API key
CREDENTIALS
Per-tenant encrypted vault
MODEL SEES
Results only, never credentials
AUDIT LOG
Every call, on every plan
05 · WORKS WITH

Use BrowserAct from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
06 · RELATED

More web data and scraping apps.

Firecrawl30 actionsTavily5 actionsExa20 actionsAgentql3 actionsAgenty79 actionsAirtop10 actionsJust One API288 actionsApify183 actions
ALL WEB DATA AND SCRAPING APPS →
07 · FAQ

BrowserAct and Arc0, answered.

Q01

Can I use BrowserAct with Claude, ChatGPT or Cursor?

Yes. Connect BrowserAct to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the BrowserAct actions you allow.

Q02

How do users connect BrowserAct?

Users add their BrowserAct api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

Q03

Which BrowserAct actions can my agent take?

11 in total: 7 read, 3 write and 1 destructive, such as “resume Task”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in BrowserAct?

Yes. Actions like “cancel Task” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.

Q05

Can my own backend call BrowserAct too?

Yes. The same BrowserAct connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug BrowserAct into your agent.

Your users connect BrowserAct once, under your brand. Your agent gets 11 actions behind your policies, with every call on the record.

Connect BrowserActBrowse all apps →

Free to build · MCP + REST · Audit log on every plan