Unverified60% confidenceFactExact time
The idea behind WebMCP is to have websites proactively expose a machine-readable capability interface, enabling AI agents to invoke webpage functionality as if calling an API.
2
Sources
60%
Confidence
Long-term
Relevance
8/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWebAI is a set of browser-oriented web standards whose core design idea is compatibility with multiple AI models, attempting to establish a universal interface at the web layer.76% similarUnverifiedWebMCP's idea is to attach structured semantic descriptions to interactive elements such as buttons and forms, and to declare their parameters and return formats using the schema specified by MCP, so that AI agents can know what can be done on a page without understanding pixel layout or DOM hierarchy.74% similarUnverifiedAgent readability requires websites not only to have semantic HTML and structured data (such as Schema.org markup), but also to provide programmatic interfaces (such as MCP and REST APIs) for agents to directly perform operations (search, add to cart, checkout).74% similarUnverifiedWith browser-side inference capabilities powered by WebAssembly and WebGL, gesture recognition can run directly in web applications without installing additional software or relying on cloud computing resources.73% similarUnverifiedCompared with writing a separate API integration for each service, browser automation has a generality advantage: as long as there is a web page, the agent can complete operations by opening the page, parsing the DOM, clicking buttons, and filling out forms, without the target system needing to provide an official API.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/813707API
curl https://kongchang.com/api/v1/knowledge/claims/813707MCP
get_claim(id=813707)