does livekit agents have webfetch tool by default
# ask-ai
a
does livekit agents have webfetch tool by default
✅ 1
t
Thanks for reaching out, Charan. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
Thanks — I’m responding to your question about whether LiveKit agents include a WebFetch tool by default. Short answer: No — LiveKit agents do not include a
WebFetch
tool by default. What you can do instead: • Use provider-specific tools that support web search/browsing (for example, the xAI Grok Voice Agent API offers a
WebSearch
tool and Gemini also provides web-capable tools). These must be explicitly added to your agent configuration. • Create your own function/tool to perform HTTP requests or fetch external data and register it with the agent (for example via Agent Builder or a custom tool definition). If you want, tell me which provider or approach you’re planning to use (xAI, Gemini, or a custom HTTP tool) and I’ll give step-by-step guidance for adding that capability to your agent. Sources: Tool definition and use | LiveKit Documentation | Introduction | LiveKit Documentation | xAI Grok Voice Agent API plugin | LiveKit Documentation | Announcing Agent Builder: a pathway to code | LiveKit | Function tool definition | LiveKit Documentation, with interactive elements
a
can it search online public urls and answer?
t
Thanks — I’m responding to your question about whether LiveKit agents can search online public URLs and answer based on that content. Short answer: Yes — LiveKit agents can search public URLs and use that content to answer, but only when you enable or add a web-capable tool. This is not automatic. How you can enable it: • Use a provider tool that supports web search/browsing (must be explicitly added to the agent): ◦ xAI Grok provides a
WebSearch
tool that can search the web and browse pages. ◦ Google Gemini offers provider tools (e.g.,
GoogleSearch
) for web searches. ◦ Azure/OpenRouter can be configured with web-search plugins where supported. • Add a custom HTTP/tool implementation: ◦ Use Agent Builder or define a function/tool that fetches content from public URLs (HTTP requests), then pass that content back to the agent for answering. ◦ Implement retrieval-augmented workflows (RAG) or custom fetch + summarization pipelines if you need structured processing of many pages. Next step I can help with • Tell me which provider you plan to use (xAI/Grok, Gemini, Azure, OpenRouter) or whether you want to build a custom HTTP fetch tool, and I’ll give step‑by‑step setup code and configuration for that option. Sources: Azure OpenAI LLM plugin guide | LiveKit Documentation | Google Gemini LLM | LiveKit Documentation | xAI Grok Voice Agent API plugin | LiveKit Documentation | OpenRouter LLM plugin guide | LiveKit Documentation | External data and RAG | LiveKit Documentation, with interactive elements