Web search tools for LangChain agents
search1api-langchain is the official Search1API integration for LangChain. It exposes web search, news search, and page crawling through the standard tool interface, so an agent picks them up the same way it picks up any other tool.
What the toolkit provides
Three tools in one call
Search1APIToolkit().get_tools() returns web search, news search, and page crawling together, ready to hand to an agent.
The official Python SDK underneath
Authentication, retries, timeouts, and API errors are handled by the maintained SDK rather than by hand-rolled request code.
Standard tool interface
Nothing framework-specific to learn. The tools follow LangChain conventions, so they compose with the agents and chains you already have.
Install and use
Create a key in the dashboard, export it, and hand the toolkit to an agent.
Install β bash
Hand the tools to an agent β python
Worth knowing
The similarly named langchain-search1api package on PyPI is an independent third-party project. search1api-langchain is the one Search1API maintains.
SEARCH1API_KEY is accepted as well as SEARCH1API_API_KEY, so the same environment works for the Search1API CLI.
Tool calls bill as ordinary API calls: 1 credit per search or news request, plus 1 for each page successfully retrieved.
What people build with it
A research agent that searches several sources and cites what it used.
A RAG pipeline that pulls live web content alongside an internal index.
A monitoring chain that watches news on a topic and summarizes daily.
FAQ
Which package should I install?
search1api-langchain. There is a separate, unaffiliated PyPI package called langchain-search1api with the words reversed; it is not maintained by Search1API.
Can I use only one of the tools?
Yes. The toolkit returns all three so you can hand them over at once, but the individual tools can be imported and passed on their own when an agent should only be able to search, or only to read pages.
Is there a TypeScript equivalent?
For JavaScript and TypeScript applications, use the official @search1api/client SDK directly and wrap it in whatever tool interface your framework expects.