Daily · 7:00
Track competitor pricing
Checks pricing pages and flags every change.
Web scraperMCP
AI web scraping agentTurn web pages into rows.
Give it a list of pages and the fields you want. Each morning it reads them, tidies what it finds, and adds new rows to your sheet.
Your first signup includes $50 in free credits.
Daily · 6:00
Collect web pages into a sheet
- Reads
- Pages you list
- Grows
- Clean new rows
- Lands in
- Google Sheets
Ask for it once, in plain words.
No flowchart to draw. Describe the job in chat. TulipFarm sets up the agent and its workflow, and shows you both before anything runs.
You
Every day at 6, open the pages listed in my Sources sheet. Pull the name, price and date from each page. Add a row for anything new and skip what I already have.
TulipFarm
I made an agent called Page collector and a daily 6:00 schedule. It reads your list in Sheets and the pages through a Custom MCP connection. Adding rows asks you first. Activate it?
ActivateEdit first
One morning, start to finish.
Every run is saved as an agent task you can open, check and rerun.
- Schedule
Every day at 6:00
Run startedStarted - Reads
Your list of pages in Sheets
25 pages - Scrapes
Each page, through your scraper connection
23 read, 2 failed - Checks
Drop rows you already have
7 new rows - Asks first
Add the new rows to the sheet
Approved - Writes
Rows in Sheets
7 rows addedCompleted
What lands in your sheet.
The same columns every day, so you can sort, filter and chart.
Collected pages| Page | Name | Price | Seen on |
|---|---|---|---|
| Page 1 | Item A | $20 | Oct 18 |
| Page 2 | Item B | $35 | Oct 18 |
| Page 3 | Item C | $12 | Oct 18 |
| Page 4 | Item D | Not found | Oct 18 |
What gets planted.
Three parts you own and can change at any time.
- Agent
Page collector
Reads each page and pulls your fields.
- Table
Collected pages
A table of rows if you prefer it to Sheets.
- Workflow
Daily 6:00
Runs each morning with your access.
Uses
- Web scraperMCPWeb scraper
- Google Sheets
Add if you use it
- Slack
- Web searchMCPWeb search
Gmail
It reads quietly. It asks before it acts.
Reads without asking
- Your list of pages in Sheets
- The pages themselves
- Rows you already have, to skip repeats
Asks you first
- Adding rows to the sheet
- Changing any existing row
- Posting a summary to Slack
It uses your own connections, so it never sees more than you can.
Grow it your way.
Flag changes
Add a run that posts to Slack when a value on a page changes.
Weekly only
Switch to a weekly schedule for pages that rarely change.
Tell you about failures
Email yourself a list of pages that did not load.
Questions, answered.
How does it read pages?
Through a Custom MCP connection. A web scraper is not built in. You connect an app with an MCP server, choose none, OAuth or an API key, and review each tool before an agent can use it.
Is scraping allowed on every site?
Not always. Check each site's terms and robots rules before you list its pages. You decide which pages to include.
What if a page changes its layout?
The agent reads the page text, not a fixed layout, so small changes are fine. If a field goes missing it writes Not found and tells you.
On request
Research anything, with sources
A question in, a cited brief out, saved to Docs.
Web searchMCP
On request
Analyze a competitor
Reads a rival's site and news, then writes a clear comparison.
Web searchMCPWeb scraperMCP
The web, tidied into rows.
Public beta, open to everyone. Your first signup includes $50 in free credits.