JavaScript - LangSearch

Use a current Node.js runtime with native fetch and AbortSignal.timeout. Select JavaScript below, save as search.mjs, replace the key, and run node search.mjs.

curl https://api.langsearch.com/v1/web-search \
  -H "Authorization: Bearer YOUR_LANGSEARCH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "How does a search API help an AI agent?",
    "count": 5,
    "contents": {"text": true},
    "freshness": "noLimit"
  }'

import json
from urllib.request import Request, urlopen

request = Request(
    "https://api.langsearch.com/v1/web-search",
    data=json.dumps({
        "query": "How does a search API help an AI agent?",
        "count": 5, "contents": {"text": True}, "freshness": "noLimit"
    }).encode(),
    headers={
        "Authorization": "Bearer YOUR_LANGSEARCH_API_KEY",
        "Content-Type": "application/json"
    }
)
with urlopen(request, timeout=30) as response:
    result = json.load(response)
print("Input tokens:", result["usage"]["input_tokens"])
print("Output tokens:", result["usage"]["output_tokens"])
for source in result["data"]["webPages"]["value"]:
    print(source["name"], source["url"])

const response = await fetch("https://api.langsearch.com/v1/web-search", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_LANGSEARCH_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    query: "How does a search API help an AI agent?",
    count: 5, contents: { text: true }, freshness: "noLimit",
  }),
  signal: AbortSignal.timeout(30000),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
const result = await response.json();
console.log("Input tokens:", result.usage.input_tokens);
console.log("Output tokens:", result.usage.output_tokens);
for (const source of result.data.webPages.value) {
  console.log(source.name, source.url);
}

Keep requests on the server

Call LangSearch from a server route or backend service. Do not embed your real key in a public browser bundle.

Select source context

const { input_tokens, output_tokens } = result.usage;
console.log({ input_tokens, output_tokens });
const sources = result.data.webPages.value.map((item) => ({
  title: item.name,
  url: item.url,
  context: item.text || item.snippet || "",
}));

Keep the URLs with the context so your model can cite its sources. For failures, inspect HTTP status and the error body before deciding whether to retry.