1. SDK
  2. Web Search

​
Overview

Web search gives your assistant access to current information from the internet. Pass web_search="Auto" on any message and the assistant decides when web search is helpful.

web_search is a per-turn parameter — pass it on every call where you want it active.

​
Non-Streaming

  • Python

  • JavaScript

  • TypeScript

import asyncio
from backboard import BackboardClient

async def main():
    client = BackboardClient(api_key="YOUR_API_KEY")

    response = await client.send_message(
        "What are the latest developments in AI this week?",
        web_search="Auto",
    )
    print(response.content)

if __name__ == "__main__":
    asyncio.run(main())

​
Streaming

  • Python

  • JavaScript

  • TypeScript

async for chunk in await client.send_message(
    "What are today's top tech headlines?",
    web_search="Auto",
    stream=True,
):
    if chunk.get("type") == "content_streaming":
        print(chunk.get("content", ""), end="", flush=True)
print()

​
Web Search + Memory

Combine web search with memory in a single message. Since memories are stored per-assistant, pass assistant_id to retrieve from the correct memory store:

  • Python

  • JavaScript

  • TypeScript

response = await client.send_message(
    "Based on my preferences, find me the latest Python framework news",
    assistant_id="your-assistant-id",
    web_search="Auto",
    memory="Auto",
)
print(response.content)

​
Options

ValueBehavior
"Auto"Assistant decides when to search the web
"off"Disabled (default)

Web search adds latency as the system retrieves and processes current information. Use Auto mode to let the assistant decide when it’s needed.