• World expands AgentKit for verified AI agents using World ID.
• AI agents can act online on behalf of verified human users.
• System aims to prevent bots while enabling trusted automation.

World is expanding access to AgentKit, its framework designed to create human-verified AI agents and allow individuals to connect those agents to a verified World ID.

The system enables AI agents to act on behalf of users across the internet while maintaining identity verification through World’s network.

The development comes as AI agents become increasingly capable of performing online tasks such as shopping, making reservations, navigating websites, and interacting with digital services.

This growing capability has created a challenge for businesses in distinguishing between agents representing real users and automated bot networks.

AgentKit is positioned as a response to that issue by linking AI agents directly to World ID, allowing websites and applications to verify when an agent is acting on behalf of a unique human.

The framework is designed to support task delegation while maintaining safeguards tied to identity verification and user control.

How AgentKit links AI agents to verified identity

To begin using AgentKit, individuals require a verified World ID, access to World App, and a supported AI agent, including tools such as Claude Code, Codex, Cursor, Hermes, or OpenClaw.

Users connect their proof of human through World’s ToolRouter interface, generate an API key, and link their AI agent within minutes.

Once connected, the agent can interact with services that support AgentKit and perform tasks on behalf of the user.

The system is designed to allow individuals to delegate digital tasks to AI agents while preserving controls tied to verified identity.

According to the framework description, this structure is intended to ensure that AI activity remains attributable to a real human user rather than anonymous or automated systems.

Demo shows real-world use case

The technology was recently demonstrated through a limited-edition release of 500 “Human in the Loop” hats available exclusively to verified World ID holders.

During the demonstration, AI agents discovered the drop, verified eligibility, navigated the storefront, and completed purchases on behalf of users while maintaining one-item-per-person limits tied to verified identities.

All 500 hats were claimed by verified individuals across multiple countries, including the United States, Germany, Japan, and the United Kingdom.

The demonstration was used to show how AI agents can execute real-world transactions while preserving identity-based constraints designed to limit abuse.

The example highlighted how businesses could allow AI agents to complete tasks on behalf of users while still preventing exploitation by bot networks.

Building a trust layer for the agent economy

As more services integrate AgentKit, World aims to create what it describes as a trust layer for an emerging agent economy.

The goal is to enable AI agents to transact and interact online while remaining accountable to the humans they represent.

The system is intended to support a growing range of use cases where AI agents operate autonomously but within a framework of verified identity and user authorization.

This includes both commercial applications and broader digital service interactions.

The World project was originally conceived by Sam Altman, Max Novendstern, and Alex Blania, and aims to provide proof of human, finance and connection for every human in the age of AI.

The company says AgentKit is part of its broader effort to support identity verification in an environment where AI agents are becoming increasingly capable of acting independently across online platforms.