Tools / Web search and scraping / API
POST a query to web.search and get the ranked results as JSON: position, title, URL, domain and a snippet for each, ready to cite or to scrape next. site:, after: and before: in the query work as filters. The call routes across several search providers and moves on when one fails, all behind one key.
Ask your agent
Search the web for open source vector databases and list the names that come up most across the results.
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 | domain | title |
|---|---|---|
| 1 | lakefs.io | Best 17 Vector Databases for 2026 [Top Picks] |
| 2 | aws.amazon.com | What is a Vector Database? - AWS |
| 7 | sqlpipe.com | Best Open-Source Vector Databases |
Nine results for open source vector database: vendor blogs, an AWS explainer, a Reddit list and community posts, each with its domain and snippet.
Step by step
One RouterGrowth key, sent as a Bearer token.
Send capability web.search with input query and a limit to /v1/run.
Each result has position, title, url, domain and snippet.
Run web.scrape on the URLs you keep to get the full page text.
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.
Search the web for open source vector databases and list the names that come up most across the results.
Search for reviews of our three closest competitors, then scrape the top two results for each and summarize the complaints.
Find what was published about our category in the last month, on the web and in the news, and split press from blogs.
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": "open source vector database", "limit": 8}'curl -X POST https://api.routergrowth.com/v1/run \
-H "Authorization: Bearer $ROUTERGROWTH_API_KEY" \
-d '{"capability": "web.search", "input": {"query": "open source vector database", "limit": 8}}'Worth knowing
Results include forums and social pages next to blogs and docs. Filter on domain if you only want some.
This run asked for eight and returned nine. Cap the list in code when the count matters.
A search takes from about a second to under a minute, depending on which provider answers. This one took 42 seconds.
FAQ
Several are wired, and the run reports which one answered. This example ran on Apify.
Yes. Add site:domain to the query or pass include_domains with up to 10 domains.
Yes. start_date and end_date remove results dated outside the range; undated results are kept and counted.
No. It returns snippets. Use web.scrape on a URL for the page text.
More on Web search and scraping
Reading this as an agent? This page as markdown: /tools/web-scraping/web-search-api.md
Connect your agent once. The first calls run on the $1 that comes with signup.