원문에서 더 깊이 읽기

GEO·AI 검색·마케팅 관련 자료를 주제별로 찾아보세요.

원문 1095건 · 7 / 46페이지

Using tools with Gemini API

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Using tools with Gemini API Tools extend the capabilities of Gemini models, enabling them to take action in the world, access real-time information, and perform complex computational tasks. Models can use tools in both standard request-response interactions and real-time streaming sessions using the Live API . Tools are specific capabilities (like Google Search or Code Execution) that a model can use to answer queries. The Gemini API provides a suite of fully managed, built-in tools, or you can define custom tools

Googlebot

Home Search Central Documentation Send feedback Googlebot Googlebot is the generic name for two types of web crawlers used by Google Search: Googlebot Smartphone : a mobile crawler that simulates a user on a mobile device. Googlebot Desktop : a desktop crawler that simulates a user on desktop. You can identify the subtype of Googlebot by looking at the HTTP user-agent request header in the request. However, both crawler types obey the same product token (user agent token) in robots.txt, and so you cannot selectively target either Googlebot Smartphone or Googlebot Desktop using robots.txt. For

Gemini Deep Research agent

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Gemini Deep Research agent The Gemini Deep Research agent autonomously plans, executes, and synthesizes multi-step research tasks. Powered by Gemini, it navigates complex information landscapes to produce detailed, cited reports. New capabilities allow you to collaboratively plan with the agent, connect to external tools using MCP servers, include visualizations (like charts and graphs), and provide documents directly as input. Research tasks involve iterative searching and reading and can take several minutes to

Mitigate clickjacking with XFO or CSP Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Mitigate clickjacking with XFO or CSP Stay organized with collections Save and categorize content based on your preferences. Clickjacking is a malicious attack where users are tricked into clicking on links or UI elements on a site that appears to be a trusted and familiar site. This is typically accomplished by embedding part or all of the trusted site into the malicious site using an . The X-Frame-Options (XFO) header and the frame-ancestors directive in the Content-Security-Policy (CSP) header can mitigate clickjacking attacks by controlling how a

Troubleshoot Google Search crawling errors

Home Search Central Documentation Send feedback Troubleshoot Google Search crawling errors Here are the key steps to troubleshooting and fixing Google Search crawling issues for your site: See if Googlebot is encountering availability issues on your site . See whether you have pages that aren't being crawled, but should be . See whether any parts of your site need to be crawled more quickly than they already are. Improve your site's crawl efficiency. Handle overcrawling of your site . See if Googlebot is encountering availability issues on your site Improving your site availability won't neces

Hooks

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Hooks Hooks let you run custom scripts or external HTTP requests right before or after the agent executes code or modifies files inside its remote sandbox. Use hooks to extend the agent loop with automated guardrails and background workflows, such as: Enforcing safety and access guardrails before high-risk shell commands or restricted file reads execute. Automating data pipeline transformations right after an agent creates or modifies files. Streaming enterprise audit telemetry to external monitoring systems after

Use a strong HSTS policy Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Use a strong HSTS policy Stay organized with collections Save and categorize content based on your preferences. Plaintext protocols, such as HTTP, can be vulnerable to eavesdropping attacks , where an attacker is able to read the transmitted content. Transport Layer Security (TLS) can encrypt the traffic and make it significantly harder for attackers to use this data if captured. However, it's possible for attackers to circumvent TLS by forcing encrypted connections to use plaintext HTTP. To address this problem, you can use the HTTP Strict Transport

How to combine sitemap extensions

Home Search Central Documentation Send feedback How to combine sitemap extensions Sitemap extensions are a great way to tell Google about the different kinds of content and their metadata that you're using on your site. Often the content on your pages may fit into multiple kinds of extensions; for example, you might be publishing news articles that embed images and videos. Additionally, your pages may be localized as well, which might mean that you could add hreflang annotations for your localized pages. Namespaces For each sitemap extension that you want to use in a sitemap you need to specif

Does not redirect HTTP traffic to HTTPS Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Does not redirect HTTP traffic to HTTPS Stay organized with collections Save and categorize content based on your preferences. All sites should be protected with HTTPS. Learn why and how to set up HTTPS on your server . How the Lighthouse HTTP redirection audit fails Lighthouse flags pages that aren't redirected to HTTPS: Lighthouse audit that shows that HTTP traffic isn't redirected to HTTPS. Lighthouse changes the page's URL to HTTP, loads the page, then waits for the Chrome Remote Debugging Protocol event that indicates that the page is secure. If

Credentials in managed agents

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Credentials in managed agents Credentials are server-managed secrets that let your agents reach third-party services without the secret ever entering the agent's environment. You store a credential once, reference it by ID, and the egress proxy resolves and injects it at request time. Secret values are write-only. Once stored, they are never returned by any endpoint, so a compromised agent cannot read back the tokens it is using. The primary place you use a credential is the network allowlist on environment.networ

Does not use HTTPS Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Does not use HTTPS Stay organized with collections Save and categorize content based on your preferences. All sites should be protected with HTTPS, even ones that don't handle sensitive data. This includes avoiding mixed content , where some resources are loaded over HTTP despite the initial request being served over HTTPS. HTTPS prevents intruders from tampering with or passively listening in on the communications between your app and your users, and is a prerequisite for HTTP/2 and many new web platform APIs. For more information about why all sites

Video sitemaps and alternatives

Home Search Central Documentation Send feedback Video sitemaps and alternatives A video sitemap is a sitemap with additional information about videos hosted on your pages. Creating a video sitemap is a good way to help Google find and understand the video content on your site, especially content that was recently added or that we might not otherwise discover with our usual crawling mechanisms. Google recommends using video sitemaps, however we also support mRSS feeds . Video sitemap best practices Video sitemaps are based on generic sitemaps, so the general sitemap best practices also apply to

Environments in managed agents

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Environments in managed agents Environments are managed Linux sandboxes that give agents an isolated place to execute code and persist files. They are decoupled from interaction context, so you can reuse the same environment across multiple interactions or start fresh at any time. The following example demonstrates how to create an interaction with a fresh remote environment and retrieve its ID: Python from google import genai client = genai . Client () interaction = client . interactions . create ( agent = "antig

Ensure CSP is effective against XSS attacks Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Ensure CSP is effective against XSS attacks Stay organized with collections Save and categorize content based on your preferences. A Content Security Policy (CSP) helps to ensure any content loaded in the page is trusted by the site owner. CSPs mitigate cross-site scripting (XSS) attacks because they can block unsafe scripts injected by attackers. However, the CSP can easily be bypassed if it is not strict enough. Check out Mitigate cross-site scripting (XSS) with a strict Content Security Policy (CSP) for more information. Lighthouse collects CSPs en

Includes front-end JavaScript libraries with known security vulnerabilities Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Includes front-end JavaScript libraries with known security vulnerabilities Stay organized with collections Save and categorize content based on your preferences. Intruders have automated web crawlers that can scan your site for known security vulnerabilities. When the web crawler detects a vulnerability, it alerts the intruder. From there, the intruder just needs to figure out how to exploit the vulnerability on your site. How this Lighthouse audit fails Lighthouse flags front-end JavaScript libraries with known security vulnerabilities: To detect vu

Building managed agents

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Building managed agents Managed agents on the Gemini API let you extend the Antigravity agent with your own instructions, skills, and data. You can customize the agent inline at interaction time, or save the configuration as a managed agent you invoke by ID. Customize the Antigravity agent The fastest way to build a custom agent is to pass your configuration inline while creating a new interaction with no registration step required. You can extend the agent in several key ways: Model selection : Choose the underly

News sitemaps

Home Search Central Documentation Send feedback News sitemaps If you are a news publisher, use news sitemaps to tell Google about your news articles and additional information about them. You can either extend your existing sitemap with news specific tags, or create a separate news sitemap that's reserved just for your news articles. Either option is fine with Google, however creating a separate sitemap just for your news articles may enable better tracking of your content in Search in Search Console. News sitemap best practices News sitemaps are based on generic sitemaps, so the general sitem

Links to cross-origin destinations are unsafe Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Links to cross-origin destinations are unsafe Stay organized with collections Save and categorize content based on your preferences. When you link to a page on another site using the target="_blank" attribute, you can expose your site to performance and security issues: The other page may run on the same process as your page. If the other page is running a lot of JavaScript, your page's performance may suffer. The other page can access your window object with the window.opener property. This may allow the other page to redirect your page to a maliciou

Antigravity agent

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Antigravity agent The Antigravity agent is a general-purpose managed agent on the Gemini API. A single API call gives you an agent that reasons, executes code, manages files, and browses the web inside your own secure Linux sandbox, hosted by Google. It is built with Gemini 3.8 Flash and uses the same harness as the Antigravity IDE. You can configure the underlying Gemini model using agent_config . Available through the Interactions API and Google AI Studio . Python from google import genai client = genai . Client

Use passive listeners to improve scrolling performance Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Get started Use passive listeners to improve scrolling performance Stay organized with collections Save and categorize content based on your preferences. Touch and wheel event listeners are useful for tracking user interactions and creating custom scrolling experiences, but they can also delay page scrolling. Currently, browsers can't know if an event listener will prevent scrolling, so they always wait for the listener to finish executing before scrolling the page. Passive event listeners solve this problem by letting you indicate that an event listener will never prevent

Managed agents quickstart

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Managed agents quickstart This guide walks you through creating and using Managed Agents on the Gemini API, using the Antigravity agent . You'll make your first agent call, continue a multi-turn conversation, stream the response, download files from the sandbox, and work with the Antigravity managed agent. Run your first agent interaction A single call to the Interactions API provisions a Linux sandbox, runs the agent loop, and returns the result. You'll define three parameters: Pass in the agent as "antigravity-p

Uses document.write() Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Get started Uses document.write() Stay organized with collections Save and categorize content based on your preferences. Using document.write() can delay the display of page content by tens of seconds and is particularly problematic for users on slow connections. Chrome therefore blocks the execution of document.write() in many cases, meaning you can't rely on it. In the Chrome DevTools Console you'll see the following message when you use document.write() : [Violation] Avoid using document.write(). In the Firefox DevTools Console you'll see this message: An unbalanced tre

Agents overview

Gemini 4 Argon is now available. Try it out . Home Gemini API Docs Send feedback Agents overview Managed agents on the Gemini API give you a configurable agent harness. A single API call provisions a Linux sandbox where the agent reasons, executes code, manages files, and browses the web autonomously. rocket_launch Quickstart Make your first agent call, stream responses, and build a custom agent. smart_toy Antigravity Agent Capabilities, tools, multimodal input, and pricing for the default agent. experiment Agents in AI Studio Visual playground for prototyping agents without writing code. Avai

Charset declaration is missing or occurs too late in the HTML Stay organized with collections Save and categorize content based on your preferences.

Home Docs Lighthouse Best Practices audits Charset declaration is missing or occurs too late in the HTML Stay organized with collections Save and categorize content based on your preferences. Servers and browsers communicate with each other by sending bytes of data over the internet. If the server doesn't specify which character encoding format it's using when it sends an HTML file, the browser won't know what character each byte represents. The character encoding declaration specification solves this problem. How the Lighthouse charset audit fails Lighthouse flags pages that do not specify th