--- title: WebMCP By Google, Clearly Explained! | daily.dev description: WebMCP is a browser API from the Chrome and Edge teams that lets websites declare their own actions (search, add to cart, checkout, etc.) as named, typed tools... canonical: https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: WebMCP By Google, Clearly Explained! | daily.dev og:description: WebMCP is a browser API from the Chrome and Edge teams that lets websites declare their own actions (search, add to cart, checkout, etc.) as named, typed tools... og:url: https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq og:image: https://api.daily.dev/og/posts/qJ3yrneMq.png og:image:alt: WebMCP By Google, Clearly Explained! og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # WebMCP By Google, Clearly Explained! **[Daily Dose of Data Science \| Avi Chawla \| Substack](https://daily.dev/sources/dailydoseofds)** · 9 min read · 0 upvotes · 0 comments ## Summary WebMCP is a browser API from the Chrome and Edge teams that lets websites declare their own actions (search, add to cart, checkout, etc.) as named, typed tools that AI agents can call directly, instead of agents guessing from screenshots or generic DOM scraping. It's compared against five other ways agents can reach an app (raw API, backend MCP server, computer use, browser automation, built-in assistant), with WebMCP argued to be the only option that requires no user setup, keeps the site's own agent, and gives structured named actions. Includes code examples for registering a tool via JavaScript and via HTML form attributes. The piece also opens with an unrelated plug for an open-source multi-model inference engine called SIE from Superlinked. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is WebMCP and how does it let AI agents interact with a website? WebMCP is a browser API from the Chrome and Edge teams that lets a website register its own actions, such as search or add-to-cart, as named tools with descriptions and typed inputs using document.modelContext.registerTool(). An agent visiting the page can then call these declared actions directly instead of interpreting screenshots or generic DOM elements, running the action inside the user's own browser session with no separate API key or login. _Developers weighing how to expose site actions to agents can track WebMCP's rollout alongside other coding tool updates on daily.dev._ ### How is WebMCP different from browser automation or computer-use agents for interacting with websites? WebMCP lets a site declare named, typed actions upfront so an agent gets structured intent directly, unlike computer use which reads pixels and clicks based on visual guesses, or browser automation which reads generic DOM structure without knowing what any given div or button actually means. WebMCP requires no user setup, keeps the agent under the visitor's control, and runs actions inside the existing browser session rather than needing a separate API key or login. _Anyone deciding between browser automation and declarative tool APIs can follow these agent-integration approaches on daily.dev._ ### How do I add a form to my website so an AI agent can use it without writing JavaScript? Add two attributes, toolname and tooldescription, directly to an existing HTML form element, and the browser reads the form's inputs to build the JSON Schema itself, exposing it as a callable action to agents with no additional JavaScript required. This works for any form already on a page, such as a flight search form with 'from' and 'to' fields. _Teams prototyping agent-ready forms can keep tabs on WebMCP implementation examples via daily.dev._ ## Similar posts on daily.dev - [WebMCP turns any Chrome web page into an MCP server for AI agents](https://daily.dev/posts/webmcp-turns-any-chrome-web-page-into-an-mcp-server-for-ai-agents-t7se7hpwy) · The New Stack · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#automation](https://daily.dev/tags/automation), [#mcp](https://daily.dev/tags/mcp), [#google-chrome](https://daily.dev/tags/google-chrome) [View this post on daily.dev](https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"WebMCP By Google, Clearly Explained!","url":"https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq"},"datePublished":"2026-08-31T13:46:08.759Z","dateModified":"2026-09-13T19:46:53.560Z","description":"WebMCP is a browser API from the Chrome and Edge teams that lets websites declare their own actions (search, add to cart, checkout, etc.) as named, typed tools...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b643017984a29762ab7525b2ddd521fa?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b643017984a29762ab7525b2ddd521fa?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Daily Dose of Data Science | Avi Chawla | Substack","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Daily Dose of Data Science | Avi Chawla | Substack","logo":"https://media.daily.dev/image/upload/s--4IHQgTOw--/f_auto/v1710503712/logos/dailydoseofds","url":"https://daily.dev/sources/dailydoseofds"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,automation,mcp,google-chrome","timeRequired":"PT9M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Daily Dose of Data Science | Avi Chawla | Substack","item":"https://daily.dev/sources/dailydoseofds"},{"@type":"ListItem","position":3,"name":"WebMCP By Google, Clearly Explained!"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/webmcp-by-google-clearly-explained--qj3yrnemq#faq","mainEntity":[{"@type":"Question","name":"What is WebMCP and how does it let AI agents interact with a website?","acceptedAnswer":{"@type":"Answer","text":"WebMCP is a browser API from the Chrome and Edge teams that lets a website register its own actions, such as search or add-to-cart, as named tools with descriptions and typed inputs using document.modelContext.registerTool(). An agent visiting the page can then call these declared actions directly instead of interpreting screenshots or generic DOM elements, running the action inside the user's own browser session with no separate API key or login. Developers weighing how to expose site actions to agents can track WebMCP's rollout alongside other coding tool updates on daily.dev."}},{"@type":"Question","name":"How is WebMCP different from browser automation or computer-use agents for interacting with websites?","acceptedAnswer":{"@type":"Answer","text":"WebMCP lets a site declare named, typed actions upfront so an agent gets structured intent directly, unlike computer use which reads pixels and clicks based on visual guesses, or browser automation which reads generic DOM structure without knowing what any given div or button actually means. WebMCP requires no user setup, keeps the agent under the visitor's control, and runs actions inside the existing browser session rather than needing a separate API key or login. Anyone deciding between browser automation and declarative tool APIs can follow these agent-integration approaches on daily.dev."}},{"@type":"Question","name":"How do I add a form to my website so an AI agent can use it without writing JavaScript?","acceptedAnswer":{"@type":"Answer","text":"Add two attributes, toolname and tooldescription, directly to an existing HTML form element, and the browser reads the form's inputs to build the JSON Schema itself, exposing it as a callable action to agents with no additional JavaScript required. This works for any form already on a page, such as a flight search form with 'from' and 'to' fields. Teams prototyping agent-ready forms can keep tabs on WebMCP implementation examples via daily.dev."}}]} ```