docs/usage/agent/web-search.mdx
LobeHub's web search integration lets Agents access current information from the internet — real-time news, prices, documentation, and facts beyond their training cutoff. When web search is active, the Agent retrieves up-to-date answers and cites the sources it used.
With web search enabled, Agents can:
When you ask a question that requires current information:
This happens automatically when the AI determines web search would be helpful, or on-demand when you explicitly request it.
When web search is used, a search grounding section appears above the AI's response as a collapsible card showing:
Expand the grounding card to see exactly where the information came from and evaluate source credibility.
Web search is not active by default — it requires plugin or MCP configuration.
<Tabs> <Tab title="Via Skills"> Install a web search Skill from the marketplace (Settings → Skills). Options include general web search, news search, and academic search. Some may require an API key from the search provider. </Tab> <Tab title="Via MCP Servers"> Connect a search MCP server (Brave Search, Google Search, DuckDuckGo, etc.) in your MCP settings. This gives more control and supports custom search implementations. </Tab> <Tab title="Per-Agent Configuration"> In Agent Settings → Skills, enable web search for that specific Agent and choose the search provider. This lets you control which Agents have web access. </Tab> </Tabs>Web search activates automatically for questions that clearly require current data:
"What's the current price of Bitcoin?"
"Who won the World Cup in 2024?"
"What are the latest developments in AI?"
You can also explicitly request a web search:
"Search the web for recent reviews of the iPhone 15"
"Look up the latest research on climate change"
"Find current mortgage rates in California"
Each search builds on the conversation context. Chain searches to go deeper:
"What are the top programming languages in 2024?"
→ "Search for job market trends for Python developers"
→ "Find salary data for senior Python engineers in Europe"
The AI generates optimized search queries tailored to your question, including adding time context, using precise terminology, and breaking complex questions into multiple targeted searches.
Results are ranked by relevance, authority, and recency. The AI considers whether sources are authoritative (news outlets, official docs, academic papers) vs. informal (blogs, forum posts).
The AI synthesizes findings from multiple sources into a coherent response, noting when sources agree or disagree, and flagging any uncertainty.
Every web search response includes citations. Check them:
Be specific — "What are the latest LLM releases in November 2024?" gets far better results than "Tell me about AI".
Include time context — Add "recent", "latest", "current", or specific dates to get timely results.
Ask for comparisons — "Compare X and Y" or "What are the pros and cons of Z" triggers useful multi-source searches.
Review the citations — Always expand the grounding card to see what sources were consulted and judge their quality.
Use follow-ups — Chain questions to go from broad overview to specific details.
<Callout type={'warning'}> Web searches may be logged by the search provider you've configured. Avoid searching for sensitive or personally identifying information. </Callout>
Search queries are sent to your configured search provider. Source websites may also track visits when you click through to citations. Review the privacy policy of your chosen search provider for details.
<Callout type={'info'}> Web search requires either a search plugin or MCP server to be configured. Some search providers require their own API key or subscription. </Callout>
<Cards> <Card href={'/docs/usage/community/mcp-market'} title={'MCP Marketplace'} /><Card href={'/docs/usage/agent/scheduled-task'} title={'Scheduled Tasks'} />
<Card href={'/docs/usage/agent/chain-of-thought'} title={'Chain of Thought'} /> </Cards>