---
title: "Hundreds of agents build a browser"
canonical: "https://agenticfeats.com/c/hundreds-of-agents-build-a-browser-cursor/"
agents: ["Cursor"]
evidence_grade: "A"
outcome: "partial"
job_type: "Coding"
roles: ["Developer"]
source_url: "https://cursor.com/blog/scaling-agents"
evidence_url: "https://github.com/wilsonzlin/fastrender"
source_date: "2026-01-14"
published: "2026-10-13"
updated: "2026-10-13"
---

> From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt

# Hundreds of agents build a browser

Cursor ran hundreds of concurrent coding agents, split into planners and workers, on one project for close to a week to build a web browser.

Result: The agents wrote over 1 million lines of code across 1,000 files, and a three-week Solid to React migration reached +266K/-193K edits.

## How it was set up

Planner agents create tasks and spawn sub-planners, workers complete tasks and push to one branch, and a judge agent decides whether to continue. GPT-5.2 models were used.

## Worth knowing

The Solid to React migration still needs careful review. Agents sometimes run too long and need periodic fresh starts to combat drift.

## Evidence note

Public GitHub repos for the browser and other experiments, plus a PR screenshot

## Sources

- Original: https://cursor.com/blog/scaling-agents
- Evidence: https://github.com/wilsonzlin/fastrender

## The receipt

- Agent: Cursor
- Evidence: Shown (grade A, artifact linked)
- Outcome: Partial
- Job type: Coding
- Roles: Developer
- Source: Blog
- Published: 2026-01-14
- Added: 2026-10-13