---
title: "Expanding our long-running agents research preview"
description: "Cursor has expanded its long-running agents research preview to all Ultra, Teams, and Enterprise users. The preview targets autonomous work on ambitious projects, using a custom harness designed to address frontier models’ failures on long-horizon tasks. Long-running agents propose a plan for approval before execution and use multiple agents to check work and maintain follow-through. Reported examples include runtimes of 36, 30, and 25 hours, while initial results showed substantially larger PRs with merge rates comparable to other agents. Internal examples include a Rust migration and custom kernels for video rendering, a 10,000-line network-policy PR, and secure sudo prompting for Cursor CLI; the company says agents can now deliver working solutions after hours or days, though safer production deployment and collaboration across agents remain active areas."
---

# Expanding our long-running agents research preview

[Cursor](https://yomu.fyi/company/cursor) · Cursor · Feb 12, 2026

**Type:** Announcement

## Summary

Cursor has expanded its long-running agents research preview to all Ultra, Teams, and Enterprise users. The preview targets autonomous work on ambitious projects, using a custom harness designed to address frontier models’ failures on long-horizon tasks. Long-running agents propose a plan for approval before execution and use multiple agents to check work and maintain follow-through. Reported examples include runtimes of 36, 30, and 25 hours, while initial results showed substantially larger PRs with merge rates comparable to other agents. Internal examples include a Rust migration and custom kernels for video rendering, a 10,000-line network-policy PR, and secure sudo prompting for Cursor CLI; the company says agents can now deliver working solutions after hours or days, though safer production deployment and collaboration across agents remain active areas.

## Context

Frontier models were found to fail predictably on long-horizon tasks: a slightly wrong assumption could lead to an incorrect solution, while models could also lose track of the broader task or stop after partial completion.

## Approach / What changed

Cursor created a custom harness for different frontier models. Long-running agents propose plans for approval before execution and use multiple agents to check each other’s work, helping them complete larger tasks with less supervision.

## Takeaways

- The research preview is available to all Ultra, Teams, and Enterprise users at cursor.com/agents.
- Example long-running tasks included building a chat platform in 36 hours, implementing a mobile app in 30 hours, and refactoring authentication and RBAC in 25 hours.
- Internal uses included migrating a video renderer to Rust with custom kernels, creating a 10,000-line network-policy pull request, and implementing secure sudo password prompting in Cursor CLI.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Rust](https://yomu.fyi/topic/rust)

- Source: [Cursor](https://cursor.com/blog/long-running-agents)
- Source URL: https://cursor.com/blog/long-running-agents
- Ingested by Yomu: 2026-09-02T11:44:26.553Z

[Read original post](https://cursor.com/blog/long-running-agents)