ClaimGate started as a solo hackathon project focused on one question:
How do you stop an AI agent from performing an irreversible action when its output is wrong?
I built the project around backend authorization boundaries, deterministic policy, evidence verification, audit/replay, and Foxit document workflows.
I completed the core system, but missed the original hackathon submission deadline while finishing the live verification and demo flow.
Rather than leaving it as an unfinished draft, I’m keeping ClaimGate public as an ongoing portfolio project and continuing to improve it.
Log in or sign up for Devpost to join the conversation.