Tools / Web search and scraping / API
An agent that answers from the web needs sources it can cite and you can trust. web.search takes a query plus include_domains (up to 10) or start_date and end_date, removes results outside them whichever provider answered, and returns title, URL, domain and snippet per result for the model to quote.
Ask your agent
Explain how PostgreSQL logical replication handles schema changes, using only postgresql.org, and cite each page you use.
Connect RouterGrowth once, then ask in plain words. New workspaces start with $1 of credit. Get a key or read the setup guide.
A real run
One call to web.search (agent web search), run through RouterGrowth on October 26, 2026. These are the first rows it returned; fields that name a person are masked.
| position | title | url |
|---|---|---|
| 1 | Documentation: 18: Chapter 29. Logical Replication | https://www.postgresql.org/docs/current/logical-replication.html |
| 2 | Documentation: 18: 26.1. Comparison of Different Solutions | https://www.postgresql.org/docs/current/different-replication-solutions.html |
| 3 | Documentation: 18: 29.8. Restrictions | https://www.postgresql.org/docs/current/logical-replication-restrictions.html |
Ten results for logical replication limited to postgresql.org, all from the PostgreSQL 18 docs; the filter removed none because the provider had applied it already.
Step by step
Connect RouterGrowth as an MCP server, or call /v1/run from your tool code.
Pass include_domains for trusted sources, or start_date for recent pages only.
Run web.scrape on the top URLs so the model quotes the page, not the snippet.
Return each claim with the url it came from, so the answer can be checked.
Price
List prices on this call, read from the catalog when this page was built. The exact figure is quoted before every run, and a monthly plan lowers it.
In practice
You describe the job; the agent picks the calls, shows the price and runs them.
Explain how PostgreSQL logical replication handles schema changes, using only postgresql.org, and cite each page you use.
Find pages on our category published since the first of this month and summarize them with links.
Connect
One MCP server for Claude, ChatGPT, Claude Code, Codex and Cursor, or the same call over the CLI and REST with the input above.
claude mcp add --transport http routergrowth https://api.routergrowth.com/mcp
npm install -g routergrowth && routergrowth setup
routergrowth run -c web.search -i '{"query": "logical replication", "include_domains": ["postgresql.org"], "limit": 6}'curl -X POST https://api.routergrowth.com/v1/run \
-H "Authorization: Bearer $ROUTERGROWTH_API_KEY" \
-d '{"capability": "web.search", "input": {"query": "logical replication", "include_domains": ["postgresql.org"], "limit": 6}}'Worth knowing
include_domains and the dates are applied to what comes back, so an off list result is removed even when a provider ignores the filter. filters in the response counts the removals.
With start_date, pages without a date are kept and counted in filters.kept_undated. Drop them when freshness matters.
A snippet is a line or two. Scrape the URL before the model quotes it.
FAQ
Yes. Connect it as an MCP server or call the REST endpoint from a tool; the agent sends the query and the filters.
Up to 10 in include_domains, and their subdomains count.
There is no exclude field. Name the domains you want, or drop results in code.
No. This run asked for six and returned ten, so trim the list before it reaches the model.
More on Web search and scraping
Reading this as an agent? This page as markdown: /tools/web-scraping/web-search-api-for-llm.md
Connect your agent once. The first calls run on the $1 that comes with signup.