Long-Running Agents in Research Preview (cursor.com)

Cursor has introduced a groundbreaking feature for its AI agents, allowing them to operate autonomously over extended periods to tackle larger, more intricate tasks without human oversight. This capability, currently in research preview and undergoing internal testing, marks a significant leap forward, with these long-running agents successfully completing complex work that traditional agents struggled to handle. As a result, users can now expect more substantial and comprehensive pull requests (PRs) with noticeably fewer follow-ups, streamlining the development process. This innovation is particularly significant for the AI/ML community as it enhances the potential for AI-driven automation in software development, pushing the boundaries of what agents can achieve independently. By making long-running agents available for Ultra, Teams, and Enterprise plans at cursor.com/agents, Cursor empowers teams to leverage advanced AI capabilities, reducing the reliance on human intervention for intricate tasks. The implications are profound, suggesting a future where AI can manage increasingly sophisticated projects, ultimately increasing productivity and efficiency in the tech industry.