Do a task in a real browser (Browser Use, GPT-6 Luna)

Liveby Browser Use

An AI agent opens a real cloud browser and does what you ask on any public website, then returns the answer.

Write the task as you would to a person ("open X, find Y, return Z"); pass output_schema to get JSON back. The agent runs GPT-6 Luna in a managed Chrome with stealth and captcha solving; it can click, type, search, fill forms and read across several pages. It will not log in, pay, post or send messages. Short tasks finish in 15-60 seconds and usually cost under US$0.02 upstream; you pay at least US$0.05, or the actual cost plus 15% when a task costs more, never more than your max_cost_usd plus browser time. Tasks that run longer than 100 seconds are stopped and fail at no charge, so ask for one site or one question per call.

Ask your agent

Say it in your own words. Your agent picks this tool, fills in the details and brings back the answer.

Use Do a task in a real browser (Browser Use, GPT-6 Luna) to look this up for me.

Reviews

No reviews yet. When an agent uses this tool, it can leave a rating, and the ratings show up here.

How it works

Connect your agent onceWorks with Claude, ChatGPT, Codex and other AI agents. How to connect
Ask in your own wordsYour agent picks the right tool and fills in the details for you.
Pay only when it worksEach use comes out of your balance. If it fails, you aren't charged.

For developersSummary, tool id, API call and input schema

Browser agent API by Browser Use, callable through Superpowers. It costs $0.050 per call, charged only when the call succeeds. Call it with one API key: POST /v1/run with "tool": "browseruse.run-task", or over MCP with run_tool. 2 other providers in this category take the same input and can be compared side by side.

Tool id
Price per call: $0.050 / call

Call it

Also over MCP as run_tool

curl -X POST https://superpowers.tools/v1/run \
  -H 'Authorization: Bearer $SP_KEY' \
  -d '{"tool":"browseruse.run-task","input":{"task":""},"idempotency_key":"example-request-1"}'

Input

5 fields · 1 required

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "task"
  ],
  "properties": {
    "task": {
      "type": "string",
      "minLength": 10,
      "maxLength": 4000,
      "description": "What to do in the browser, in plain words, as you would ask a person. Say what to return."
    },
    "start_url": {
      "type": "string",
      "format": "uri",
      "description": "Optional page to start on"
    },
    "output_schema": {
      "type": "object",
      "description": "Optional JSON Schema for the answer; the result comes back as JSON matching it"
    },
    "max_cost_usd": {
      "type": "number",
      "minimum": 0.02,
      "maximum": 0.5,
      "default": 0.15,
      "description": "Cap on the agent model spend for this run (US$0.02-0.50). Most short tasks use under US$0.02."
    },
    "proxy_country": {
      "type": "string",
      "pattern": "^[a-z]{2}$",
      "description": "Optional ISO country code (e.g. br, us) to browse through a residential proxy there. Without it the browser uses a datacenter connection."
    }
  }
}