Technology ❯ OpenAI ❯ GPT Models
Capabilities Model Improvements AI Agents
The invite-only tool uses LLM reasoning to confirm exploitable bugs in sandboxes before proposing developer-reviewed patches.
