How to use Cursor to build an experimental browser
Willison independently built Cursor’s agent-generated FastRender browser on macOS and found a working window with legible pages and visible rendering problems.
AddedBuilding, testing, debugging and maintaining software or infrastructure, with code review and checks before changes reach users.
424 use casesShowing 73–96
Preparing search…
Willison independently built Cursor’s agent-generated FastRender browser on macOS and found a working window with legible pages and visible rendering problems.
AddedCursor replaced a flat shared-file coordination design with planners, workers and a judge, then ran large coding experiments including a browser and framework migration.
AddedDevClaw uses OpenClaw to run issue-based DEV and QA loops for isolated Telegram projects, reusing sessions and selecting models by task.
AddedThe author built RecipeNinja using Windsurf, later adding Claude Code and Gemini help, with a voice assistant that navigates the app during conversation.
Added WindsurfThe author reports a deployed Telegram Ads SaaS with 24 interactive HTML pages, 60+ API endpoints, 311 migrations, Kubernetes infrastructure and Stripe and PayPal billing.
AddedClaude identified a bad IX-Denver peering connection between the ISP and Fastly; the author later contacted the ISP and says it confirmed the diagnosis.
AddedThe author reports successful trials with mostly standard CRUD and REST API apps, while describing the generated code as messy and noting unresolved hardening and maintenance issues.
AddedThe author reports a full-stack NFL fantasy analytics platform, 58.4% in-range accuracy across 81K fantasy predictions and 4.54 MAE.
AddedThe author reports deploying PodGen.io after about three weeks of Claude conversations, with credit pricing, podcast generation, distribution integration and a user dashboard.
AddedThe author reports a working Figma binary-protocol decoder used daily for several days, supporting scenegraph extraction, clean SVG exports and design-to-HTML mapping.
AddedChase reports a fully functional Laravel poker tracking application after two or three days of Cursor conversations, including a dashboard, statistics, imports and hand viewing.
AddedThe shared Cursor demonstration identifies u.nmee as a typo for u.name, edits server.js and restarts the development server; the terminal output reports the error fixed.
AddedThe team released shouldiuse.dev after an internal AI-assisted experiment, providing public GitHub dependency checks, stored results, generated summaries and authenticated MCP access.
AddedCanary generates and runs user-flow tests from code and pull-request changes, then posts results and recordings. Its founders reported catching an invoice discrepancy before release.
Added CanaryClaude Code implemented a Go proxy exposing an HTTP request/callback interface over a WebSocket-only service, then tested it live and fixed review findings.
AddedClaude Code built a menu-bar clipboard manager through several iterations and implemented a lightweight native markdown editor from a GPT 5.5 outline.
AddedScrapAI delegates site analysis and extraction-rule setup to an agent, then uses validated JSON configurations for recurring Scrapy collection without per-page AI.
AddedWesner used Claude Code to document Micropolis, published the documents and prompt, and converted them into audio. Don Hopkins reviewed the documentation and praised specific coverage.
AddedCodev combines git-tracked specifications, enforced phase transitions, isolated builder worktrees, and Claude, Gemini, and Codex reviews. Its author reported 20 bugs caught before shipping.
AddedA captured HAR file became OpenAPI documentation, followed by a local Python MCP server that used Playwright to obtain credentials before calling internal APIs.
AddedClaude completed overdue upgrades on a branch while the author checked in occasionally. The team then performed several engineer days of manual testing and deployed.
AddedAn author used Claude Code to clean up a routing app built with Antigravity, then live-tested its shadow-aware routes and identified missing tree data.
AddedClaude Code helped refactor a C# reporting DLL, add dependency injection and tests, and create a CLI; the developer repaired logic errors before final release testing.
AddedA React developer reviewed an initial Claude Code refactor before expanding it to hundreds of edits; most worked, but declaration errors required manual repairs.
Added