03. clawguardianhackathon winner · hookemhacks '26
ClawGuardian
prompt-injection firewall for ai agents with on-chain threat sharing.
winner at hookemhacks '26. a real-time prompt-injection firewall that sits between user input and ai agents, detecting and blocking injection attacks before they reach the model. combines on-chain threat sharing so every instance learns from attacks seen by any other node in the network.