Products & Announcements

Claude Code on the Web: A Solid v1 That Outshines Cursor

AI Coding AgentsDeveloper ToolingVibe CodingAI & Productivity
Added Oct 28, 2025
Link114
Article: PositiveCommunity: PositiveMixed

The author has been using Claude Code on the web extensively and finds it effective despite its early-stage feel. It supports async development workflows by creating branches, PR-based diffs, and a teleport command to continue locally, and it works well via the iOS app. Compared to Cursor’s similar feature, Claude’s greater stability and polish make it the better experience.

Key Points

  • Claude Code on the web is a simple v1: start a thread, it runs in a container, creates a branch, and you view changes via a PR.
  • You can continue locally by using a claude --teleport <uuid> command that syncs the branch and preserves the thread.
  • The author uses it as a self-updating to-do list across multiple projects, returning later to review and ship changes.
  • iOS app availability enables on-the-go questions and answers that wait for the user.
  • Compared to Cursor’s earlier similar feature, Claude Code feels more stable and polished; product quality drives the author’s preference.

Sentiment

Mixed but generally positive toward Claude Code as a product, while acknowledging serious competition from OpenAI's Codex. The community appreciates Claude Code's polish and collaborative approach but many admit Codex produces higher quality code output. The web version gets credit for mobile use cases but criticism for limited environments. The overall tone is constructive comparison rather than dismissal.

In Agreement

  • Claude Code web enables coding from mobile devices, iPads, and anywhere away from a desk — the self-executing to-do list workflow is genuinely useful
  • The teleport feature that bridges web sessions to local CLI is a standout capability
  • Claude Code feels more polished and reliable than Cursor's web agent
  • Claude's collaborative, incremental approach to problem-solving is better suited for experienced developers who want to stay in the loop
  • Tips for customizing Claude Code via CLAUDE.md, AGENTS.md, and settings.json show a mature and configurable system

Opposed

  • GPT-5 Codex produces higher quality, more reliable code output than Claude's Sonnet models — working code on the first try
  • Claude Code web's limited environment (no devcontainers, no Docker, limited language support) makes it unusable for serious iterative development
  • The container to clone to change to PR loop is too slow for anything beyond trivial one-shot tasks
  • Claude CLI has significant bugs including memory leaks, CPU usage, tool call errors, and infinite loops that hurt the experience
  • Claude Code web creates public branches on public repos without user authorization, unlike Codex which lets you choose when to push