LangChain

StructuredTool wrappers for any LangChain / LangGraph agent, scrape, search, extract, and enrich, all callable by an LLM.

⚠️

Not published. @superscraper/langchain does not exist on npm, so there is no install command for it today. The source lives in this repository at packages/langchain-superscraper; the code below describes that source. Until it ships, wrap the REST API in your own StructuredTool, a POST /v1/scrape with the x-api-key header is the entire dependency.

Running it from source

Clone the repository and reference packages/langchain-superscraper as a workspace or file dependency. It expects @langchain/core and zod as peers.

Scrape in 5 lines

import { createScrapeTool } from '@superscraper/langchain';
 
// Reads SUPERSCRAPER_API_KEY from the environment.
const scrape = await createScrapeTool();
const result = await scrape.invoke({ url: 'https://example.com' });
console.log(result); // clean markdown + listing + provenance

In an agent

import { createAllTools } from '@superscraper/langchain';
import { createReactAgent } from '@langchain/langgraph/prebuilt';
import { ChatAnthropic } from '@langchain/anthropic';
 
const tools = await createAllTools();
const agent = createReactAgent({
  llm: new ChatAnthropic({ model: 'claude-sonnet-4-6' }),
  tools,
});
 
const result = await agent.invoke({
  messages: [{ role: 'user', content: 'What is the tech stack of stripe.com?' }],
});

Tools

Factory	Tool name	Endpoint
`createScrapeTool()`	`superscraper_scrape`	`/v1/scrape`
`createSearchTool()`	`superscraper_search`	`/v1/search`
`createExtractTool()`	`superscraper_extract`	`/v1/extract`
`createEnrichTool()`	`superscraper_enrich`	`/v1/enrich/website`
`createAllTools()`	all of the above

Claude Code Plugin LlamaIndex