World and Coinbase Debut Developer Toolkit to Solve the AI Agent ‘Trust Gap’

World has launched the beta version of a developer toolkit that enables AI agents to provide cryptographic proof of human identity through World ID.

Agentkit enables verified individuals to delegate their World ID to their AI agents. This creates a “human-backed agent” that can cryptographically prove its legitimacy to websites and APIs without revealing the user’s identity.

While previous solutions like the x402 protocol used micropayments to limit bot traffic, payments alone cannot prove uniqueness; a single bad actor could still fund thousands of bots. Agentkit solves this by providing a “proof of unique human” signal. This allows platforms to set limits based on the number of real people involved rather than the number of active agents.

The toolkit is expected to transform several high-friction online industries. In ticketing and reservations, platforms can ensure that concert tickets or dinner tables are secured by real fans rather than scalpers deploying automated networks. For digital services, free trials can be managed to prevent abuse by allocating access per unique human rather than per email or digital wallet.

“The key challenge is separating legitimate, human-backed agents from bot swarms,” said DC Builder, a research engineer at World Foundation. “Proof of human addresses this gap by allowing websites to verify that an agent represents a unique person without revealing who that person is.”