An OpenAI deployment engineer introduces evidence-based completionAgent completion verification checks evidence that an agent achieved the requested outcome rather than merely producing intermediate work., execution boundariesAn agent permission boundary defines and enforces which resources and actions an automated agent is allowed to use. and layered reviewCode review examines proposed source changes to assess their correctness, clarity and suitability before or around integration.. Sonar developer content engineer Killian Carlsen-Phelan then demonstrates a SonarQube integration with Codex.
The workshop uses a deliberately vulnerable sampleSecurity testing evaluates whether a system's protections hold under defined conditions and identifies weaknesses that need fixing. to illustrate secret detectionStatic analysis examines code or files for defects and risks without executing the program being analyzed., issue listing and a SQL injection fix. A participant raises a possible hook bypass, and the presenter acknowledges that the hook has limits.
Watch on YouTube
