Fetch Page
User wants to fetch and extract content from a web URL using Gemini CLI's native browsing capability.
Instructions
- •
Parse the user's input to extract the URL and optional instruction.
- •The first argument is the URL (must be a valid URL)
- •Everything after the URL is the extraction instruction (optional)
- •If no URL is provided, ask the user for the target URL
- •
Call the MCP tool
mcp__all-agents-mcp__fetch_pagewith:- •
url: the target URL - •
instruction(optional): extraction/focus instruction from the user - •
model(optional): defaults togemini-3-flash-previewfor speed
- •
- •
Display the response as-is. The tool returns formatted markdown with the fetched content.
Examples
- •
/all-agents-mcp:aa-fetch https://docs.example.com/api - •
/all-agents-mcp:aa-fetch https://example.com/pricing extract only the pricing table - •
/all-agents-mcp:aa-fetch https://github.com/repo/README.md summarize the key features