Skip to main content
Cursor can operate your lobstr.io account directly through the lobstr.io MCP server. You ask Agent for the data you need, and it picks the right scraper, launches the run in your account, and brings the results back into the chat or into a file in your project. Endpoint: https://mcp.lobstr.io/mcp
This is the operational server. It acts in your account and spends credits. A separate public server at https://docs.lobstr.io/mcp only answers questions about the lobstr.io API and can’t touch your account. You can add both, see Add the docs server too.

​
Before you start

  • A lobstr.io account with credits. Launching a run spends credits from your balance.
  • The Cursor desktop app.
You don’t need your API key. The connection uses OAuth: you approve it on lobstr.io’s own site, so no password ever reaches Cursor.

​
Connect Cursor to lobstr.io

1

Open your MCP config

Cursor reads MCP servers from a JSON file. Pick where the server should be available:
  • ~/.cursor/mcp.json for every project on your machine
  • .cursor/mcp.json in a project’s root for that project only
Create the file if it doesn’t exist.
2

Add the lobstr.io server

Paste this in, or add the lobstr entry to the mcpServers you already have:
Include the trailing /mcp. A URL without it won’t connect.
No client ID or secret is needed. Cursor registers itself with lobstr.io automatically.
3

Approve the connection

Cursor opens the lobstr.io consent page in your browser. You’re already signed in, so review the permissions and click Allow, then go back to Cursor.If the browser doesn’t open, open Customize in Cursor’s sidebar, find lobstr in your MCP servers, and start the sign-in from there.
4

Check it works

In Agent chat, ask “Who am I on lobstr.io?”. Agent replies with the account you’re signed in as and your plan.

​
Tool approval

By default Cursor asks before it uses any MCP tool. Click the arrow next to the tool name to see exactly what it’s about to send, such as the scraper and the inputs for a run. Cursor applies the same run modes to MCP tools as to terminal commands, so you can let read-only tools like search_scrapers, get_run, and get_results run without asking. Keep run_scraper on approval if you want to see every launch.

​
What you’re approving

The consent page lists the permissions Cursor is asking for, and you choose which ones to allow. You can revoke access at any time from your lobstr.io account.

​
What Cursor can do

New scrapers work automatically. Cursor reads each scraper’s live input schema, so nothing needs updating when lobstr.io adds one to the store.

​
Scrapers that need a synced account

Some scrapers log in to a platform on your behalf, such as the LinkedIn and Sales Navigator scrapers. They need a synced account first, set up with Account sync.
  • One matching account: if exactly one healthy synced account of the right type exists, Cursor attaches it for you.
  • Several or none: Cursor asks you which account to use, or tells you which platform to sync.
  • The type must match. A LinkedIn account can’t run a Sales Navigator scraper, and the reverse.
  • Attaching only adds. An account already on a Squid is never swapped out.

​
A typical conversation

Because Agent works inside your project, it can put the data where your code expects it.
1

Find a scraper

“Find a lobstr.io scraper for Google Maps business leads.”
2

Check what it needs

“What inputs does it take, and what does it cost?”
3

Launch the run

“Run it for dentists in Manchester.” Cursor asks you to approve the tool call, and lobstr.io asks you to confirm the cost when the estimate is high.
4

Follow the run

“Is it finished?”
5

Put the data to work

“Download the results as CSV into data/dentists.csv, then write a script that removes rows without a phone number.”

​
Add the docs server too

If you’re writing code against the lobstr.io API, add the public docs server next to the operational one. Agent can then look up endpoints and parameters while it writes your integration. It needs no sign-in.

​
Credits and confirmation

lobstr.io prices per result row, so the final cost of a run isn’t known until it finishes. Before launching, Cursor estimates the cost, and when the estimate is high or can’t be calculated, it asks you to confirm first. Reading scrapers, runs, and results never spends credits. For a closer figure before anything launches, ask Cursor to estimate the run once the Squid has its tasks. If email verification is on, it’s billed separately after the scrape.
Runs launched from Cursor are real runs in your account. They appear in your dashboard with an mcp badge next to the Squid name, count against your slots, and draw down the same credit balance as runs you launch yourself.

​
Troubleshooting

Check the URL is exactly https://mcp.lobstr.io/mcp, over HTTPS and with the /mcp path included. For details, open the Output panel (Cmd+Shift+U on Mac, Ctrl+Shift+U on Windows) and pick MCP Logs from the dropdown.
Open Customize in the sidebar and check the lobstr server is switched on. Disabled servers don’t appear in chat. You can also name the tool you want, for example “use search_scrapers to find…”.
That’s expected. lobstr.io charges per row, so the total is unknowable up front. The estimate still shows you the per-row rate.
Open the run in your lobstr.io dashboard. Cursor sees the same status your account does, so the dashboard will show the same stop reason.
Your plan caps the total slots your active Squids can reserve. The cap is checked when a Squid is created, activated, or has its slots raised, not when a run starts. If creating a Squid fails because you’re at the limit, deactivate Squids you don’t need from the dashboard, or ask Cursor to do it. After a plan change the count can read above your limit; Squids that already exist keep running.
On the free plan, reading results through the MCP server is capped at the first 30 rows. Upgrading your plan is the only way past it. See Manage subscription.