AI Coding AssistantWeb Search / Web Fetch
This article describes how to use the Web Search tool and the Web Fetch tool
Langcli comes with two built-in tools: web_search and web_fetch. To use these tools, you need to use the LangRouter api-key in Langcli, and your LangRouter account must have sufficient balance.
Each web search (web_search) costs $0.006, and each web page read via web_fetch costs $0.001.
Web Search
Including terms like "web search" or "WebSearch tool" in your prompt will trigger Langcli to use the web search tool. Here are some examples:
- Please web search for the latest news about 2026 Iran war
- Please use the WebSearch tool to find news about Trump
Web Fetch
Including terms like "read webpage" or "WebFetch tool" in your prompt will trigger Langcli to use the web fetch tool to scrape web pages and extract markdown content. Here are some examples:
- Please read the web page https://langcli.com/docs/chrome_use and then provide a summary.
- Please use the WebFetch tool to fetch https://langcli.com/docs/chrome_use and then provide a summary.
Note: If necessary, after calling the web search tool, Langcli will automatically call the WebFetch tool to read the webpage content corresponding to the URLs in the search results.
Last Updated
Chrome Use
Chrome Use — Browser Automation Quick Guide
Custom Model Providers
Langcli allows you to configure custom model providers through the modelProviders setting in your settings.json. This enables you to switch between different AI models(includes models from LangRouter and your own custom models.) using the /model command.