Web fetch tool (`web_fetch`)

The web_fetch tool allows the Gemini agent to retrieve and process content from specific URLs provided in your prompt.

Technical reference

The agent uses this tool when you include URLs in your prompt and request specific operations like summarization or extraction.

Arguments

• prompt (string, required): A request containing up to 20 valid URLs (starting with http:// or https://) and instructions on how to process them.

Technical behavior

• Network access: Destination URLs and resolved IP addresses are validated to restrict access to private, reserved, loopback, and internal networks (e.g., RFC 1918 addresses and internal domains). Transport connections are pinned to the resolved destination IP address.
• Confirmation: Triggers a confirmation dialog showing the converted URLs.
• Plan Mode: In Plan Mode, web_fetch is available but always requires explicit user confirmation (ask_user).
• Processing: Uses the Gemini API’s urlContext for retrieval.
• Fallback: If API access fails, the tool attempts to fetch raw content directly from your local machine (with destination IP validation and connection pinning applied).
• Formatting: Returns a synthesized response with source attribution.

Use cases

• Summarizing technical articles or blog posts.
• Comparing data between two or more web pages.
• Extracting specific information from a documentation site.

Next steps

• Follow the Web tools guide for practical usage examples.
• See the Web search tool reference for general queries.