I designed and built Sentinel Evidence Gate, including its trust architecture, deterministic verifiers, GPT-5.6 integration, GitHub Action, tests, and demo. It builds on my research into evidence-grounded, trustworthy AI systems.
Sentinel Evidence Gate
A trust boundary for AI-generated code: GPT-5.6 proposes security claims, while deterministic verifiers bind evidence to the exact commit and block unsafe pull requests.
Log in or sign up for Devpost to join the conversation.