An open protocol for agent feedback (feedback.now)

A new open protocol for agent feedback has been unveiled, aiming to streamline communication between AI agents and developers. This structured feedback system allows agents to discover developers' policies, submit bug reports, and send friction signals through a simple HTTP method, eliminating the need for proprietary SDKs and reducing vendor lock-in. By publishing a discovery document, developers can easily expose their endpoints, enabling any compliant agent to interact with their system. This protocol is significant for the AI/ML community as it fosters interoperability and transparency, making it easier for agents to provide valuable insights into their performance and user interactions. With three key endpoints, the process is designed for efficiency: agents learn the policy from a standard location, submit feedback via a dedicated API, and can poll for receipts to ensure their reports are received and addressed. This initiative not only enhances the feedback loop but also encourages wider adoption of collaborative AI practices, ultimately leading to improved agent reliability and user experience.