Cursor Launches Automations: AI Coding Agents That Run Without You
Cursor, the AI-powered code editor that has become a staple in the agentic coding space, has launched a new feature called Automations. The tool lets users automatically launch AI coding agents triggered by codebase changes, Slack messages, or simple timers — removing the need for engineers to constantly prompt and monitor agents.
Beyond Prompt-and-Monitor
The core idea behind Automations is breaking engineers free from the "prompt-and-monitor" cycle that defines most AI-assisted coding today. Instead of a human initiating every agent action, Cursor's framework launches agents automatically and loops humans in only when they're needed.
"It's not that humans are completely out of the picture," said Jonas Nelle, Cursor's engineering chief for asynchronous agents. "It's that they aren't always initiating. They're called in at the right points in this conveyor belt."
Hundreds of Automations Per Hour
Cursor estimates it now runs hundreds of automations per hour, far beyond simple code review. The system handles incident response through PagerDuty integrations, where an agent automatically queries server logs via MCP connections. Another automation delivers weekly summaries of codebase changes to Slack.
The feature builds on Bugbot, a long-standing Cursor tool that reviews code for bugs every time an engineer commits. Automations expand that concept into more thorough security audits and deeper code analysis. "This idea of thinking harder, spending more tokens to find harder issues, has been really valuable," said engineering lead Josh Ma.
The $2 Billion Revenue Machine
Cursor's launch comes amid fierce competition in the agentic coding space, with OpenAI and Anthropic both making significant updates to their coding tools recently. Ramp data shows Cursor holding steady with roughly 25% of generative AI clients subscribing in some capacity.
The overall growth has been staggering. Bloomberg recently reported Cursor's annual revenue has grown to more than $2 billion, doubling over the past three months alone.
The Bottom Line
Cursor is betting that the future of software engineering isn't about better prompts — it's about making agents autonomous enough that engineers can focus on architecture and decisions rather than babysitting AI. With revenue doubling every quarter and hundreds of automations running hourly, they might be right. The question is whether this makes software engineers more productive or just makes fewer of them necessary.